Get 20M+ Full-Text Papers For Less Than $1.50/day. Subscribe now for You or Your Team.

Learn More →

Immediate gain is long-term loss: Are there foresighted decision makers in the Iowa Gambling Task?

Immediate gain is long-term loss: Are there foresighted decision makers in the Iowa Gambling Task? Background: The Somatic Marker Hypothesis suggests that normal subjects are "foreseeable" and ventromedial prefrontal patients are "myopic" in making decisions, as the behavior shown in the Iowa Gambling Task. The present study questions previous findings because of the existing confounding between long-term outcome (expected value, EV) and gain-loss frequency variables in the Iowa Gambling Task (IGT). A newly and symmetrically designed gamble, namely the Soochow Gambling Task (SGT), with a high-contrast EV between bad (A, B) and good (C, D) decks, is conducted to clarify the issue about IGT confounding. Based on the prediction of EV (a basic assumption of IGT), participants should prefer to choose good decks C and D rather than bad decks A and B in SGT. In contrast, according to the prediction of gain-loss frequency, subjects should prefer the decks A and B because they possessed relatively the high-frequency gain. Methods: The present experiment was performed by 48 participants (24 males and 24 females). Most subjects are college students recruited from different schools. Each subject played the computer version SGT first and completed a questionnaire for identifying their final preference. The IGT experimental procedure was mostly followed to assure a similar condition of decision uncertainty. Results: The SGT experiment demonstrated that the prediction of gain-loss frequency is confirmed. Most subjects preferred to choose the bad decks A and B than good decks C and D. The learning curve and questionnaire data indicate that subjects can not "hunch" the EV throughout the game. Further analysis of the effect of previous choice demonstrated that immediate gain increases the probability to stay at the same deck. Conclusion: SGT provides a balanced structure to clarify the confounding inside IGT and demonstrates that gain-loss frequency rather than EV guides decision makers in these high-ambiguity gambles. Additionally, the choice behavior is mostly following the "gain-stay, lose-randomize" strategy to cope with the uncertain situation. As demonstrated in SGT, immediate gain can bring about a long-term loss under uncertainty. This empirical result may explain some shortsighted behaviors in real life. Page 1 of 10 (page number not for citation purposes) Behavioral and Brain Functions 2008, 4:13 http://www.behavioralandbrainfunctions.com/content/4/1/13 "... the brains of the normal subjects were gradually learning to Background Studies in behavioral decision-making and affective neu- predict a bad outcome, and were signaling the relative badness roscience have found that typical decision makers are fre- of the particular deck before the actual card-turning." quently "myopic" [1-5] to long-term outcome (expected (Damasio, 1994, p 220). value, EV) [6]. Conversely, Damasio [7] and Bechara et al. [8,9] proposed the Somatic Marker Hypothesis and con- In short, the Somatic Marker Hypothesis suggests that ducted the Iowa Gambling Task (IGT) to test whether ven- somatic markers, which are processed implicitly, can facil- tromedial prefrontal patients are shortsighted in terms of itate decision makers in making advantageous decisions the future and long-term outcome, and whether typical [9] and guiding explicit decision making [33,34]. decision makers can predict or foresee the future. For comparison, the notion of shortsighted vs. foresighted is The most compelling empirical evidence for supporting adopted in this study. A marked difference exists between the Somatic Marker Hypothesis is found in the IGT [8]. shortsightedness and foresight. The IGT was the central Over the past decade, the IGT has been widely employed test for a verification of the Somatic Marker Hypothesis. as a neuropsychological research instrument for investi- Some studies have attempted to replicate or modify the gating affective and executive function. At least 100 scien- protocol used in the IGT [10-12]; whereas others have tific studies have utilized the IGT to investigate a diverse proposed that the hypothesis was theoretically inade- set of neurological and psychiatric populations [35]. quate [1,5,13-16]. Lin et al. [17] pointed out that there are increasing number of studies [18-28] to demonstrate a In the IGT, card decks A and B were designated "bad" contradictory phenomenon, namely, the "prominent decks with low EV ($ -250), and C and D were "good" deck B phenomenon" [29]. The phenomenon showed decks with high EV ($ +250) in average 10 trials. Subjects that normal decision makers can not prevent their prefer- and ventromedial prefrontal patients selected "bad" decks ence to "bad" (EV) deck B in the standard version of IGT during the first 30 card-turning trials (out of 100) [7,36]; (or, due to the effect of "gain-loss frequency"). Recently, however, normal subjects gradually shifted to "good" Bechara (Sevy et al.) [30,31] also revealed a "prominent decks and avoided the "bad" decks [8,9,36-39]. However, deck B phenomenon". In their study, normal subjects pre- a careful examination of the IGT reveals a critical con- ferred the bad deck B rather than the other three decks. founding between EV and gain-loss frequency. On aver- Furthermore, the chosen number of deck B (31) in Sevy et age, for each 10-card unit, deck A contains 5 gains and 5 al. study [30] is almost a double of Bechara et al. data in losses, and deck B contains 9 gains and 1 loss. On the 1994 (about 17) [8]. On the other hand, Lin et al. [29] other hand, deck C contains 6.25 gains, 2.5 standoffs and analyzed the existing experimental results adopting sim- 1.25 losses, and D contains 9 gains and 1 loss. Altogether ple version of IGT and suggest that the "prominent deck B the bad decks (A and B) contain 14 gains and 6 losses, phenomenon" may be due to a confounding from gain- whereas good decks (C and D) contain 15.25 gains, 2.5 loss frequency. Additionally, Chiu and Lin [32] utilize a standoffs, and only 2.25 losses. Both good and bad decks modified version of IGT to demonstrate that subjects' have a similar number of gains, whereas the good decks preference to deck C is also due to gain-loss frequency, not have significantly fewer losses (see Tables 1 and 2). There- EV. A fundamental structural flaw exists in a failure of fore, it is not clear whether subjects' choices of good decks orthogonal separation between EV and gain-loss fre- in the IGT were driven by improved EV or gain-loss fre- quency (or, frequency of punishments and rewards). This quency. This study attempts to generate a symmetrical and limitation has not been properly evaluated in IGT litera- fair experiment to demonstrate the relative guiding power ture. The study tries to explore the implications of this of gain-loss frequency and EV under uncertainty, specifi- flaw. cally to identify which factor most comprehensively dom- inates normal subject preferences. To differentiate Damasio's [7] Somatic Marker Hypothesis proposes that between the relative contributions of EV and gain-loss fre- normal decision-making is often assisted by somatic quency, this study applies a new task, namely, the Soo- markers. Ventromedial prefrontal patients are influenced chow Gambling Task (SGT) [40,41] (see Table 1). largely by immediate reinforcement and are insensitive to, or cannot see, future consequences due to a lack of past Method affective experiences. Normal subjects with intact somatic A two-phase study was designed to explore learning proc- markers benefit from repeated exposure to punishments ess and final preferences during decision making. Four and rewards when performing tasks and are cognizant of card decks, randomized into 24 arrangements (e.g., future outcome. Somatic states serve as a neural expres- ABCD, BCDA, CDAB, etc.), were presented on a computer sion biasing the brain process when evaluating the bad- screen. Forty-eight college students and adults (24 males, ness or goodness of each decision. According to Damasio 24 females; age: 18–30 years, mean age: 20.71 years) par- [7]: ticipated in this study. Each subject undertook one set of Page 2 of 10 (page number not for citation purposes) Behavioral and Brain Functions 2008, 4:13 http://www.behavioralandbrainfunctions.com/content/4/1/13 Table 1: The immediate net value of each trial and gain-loss structure in the original IGT and SGT IGT Serial Numbers AB C D SGT AB C D 1 100 100 50 50 1 200 100 -200 -100 2 100 100 50 50 2 200 100 -200 -100 3 -50 100 0 50 3 200 100 -200 -100 4 100 100 50 50 4 200 100 -200 -100 5 -200 100 0 50 5 -1050 -650 1050 650 6 100 100 50 50 6 200 100 -200 -100 7 -100 100 0 50 7 200 100 -200 -100 8 100 100 50 50 8 200 100 -200 -100 9 -150 -1150 050 9 200 100 -200 -100 10 -250 100 0 -200 10 -1050 -650 1050 650 11 100 100 50 50 11 200 100 -200 -100 12 -250 100 25 50 12 200 100 -200 -100 13 100 100 -25 50 13 200 100 -200 -100 14 -150 -1150 50 50 14 200 100 -200 -100 15 -100 100 50 50 15 -1050 -650 1050 650 16 100 100 50 50 16 200 100 -200 -100 17 -200 100 25 50 17 200 100 -200 -100 18 -50 100 -25 50 18 200 100 -200 -100 19 100 100 50 50 19 200 100 -200 -100 20 100 100 0 -200 20 -1050 -650 1050 650 21 100 -1150 50 50 21 200 100 -200 -100 22 -200 100 50 50 22 200 100 -200 -100 23 100 100 50 50 23 200 100 -200 -100 24 -250 100 0 50 24 200 100 -200 -100 25 100 100 25 50 25 -1050 -650 1050 650 26 -100 100 0 50 26 200 100 -200 -100 27 -150 100 50 50 27 200 100 -200 -100 28 -50 100 50 50 28 200 100 -200 -100 29 100 100 -25 -200 29 200 100 -200 -100 30 100 100 0 50 30 -1050 -650 1050 650 31 -250 100 50 50 31 200 100 -200 -100 32 -100 -1150 50 50 32 200 100 -200 -100 33 -150 100 50 50 33 200 100 -200 -100 34 100 100 25 50 34 200 100 -200 -100 35 100 100 25 -200 35 -1050 -650 1050 650 36 100 100 50 50 36 200 100 -200 -100 37 -50 100 -25 50 37 200 100 -200 -100 38 -200 100 50 50 38 200 100 -200 -100 39 100 100 0 50 39 200 100 -200 -100 40 100 100 -25 50 40 -1050 -650 1050 650 Final Outcomes -1000 ($) -1000 ($) +1000 ($) +1000 ($) Final Outcomes -2000 ($) -2000 ($) +2000 ($) +2000 ($) Note. In these gambling tasks, the internal gamble structure of IGT and SGT and the trial to end the game is unknown to the subjects. The subjects are asked to minimize monetary expenditure and maximize their winnings by choosing one card from the four decks in each trial. In the table, negative values are marked with a bold font size. (Left part: IGT) Deck A contains the relative high-frequency loss, while decks B, C, and D contain the high-frequency gain (net value within each trial). Decks A and B have negative net value (namely, the EV), -250($) over an average of ten trials; moreover, C and D have a positive net value of +250($) over an average of ten trials. (Right part: SGT) Five cumulative trials are repeated for each deck in the Soochow Gambling Task, decks A and B have high-frequency gain, while decks C and D exhibit a reversed gain-loss pattern. The task enlarges the difference between positive and negative EVs to make the difference more noticeable than in the Iowa Gambling Task. While playing this game, subjects only experienced a gain or a loss during each trial, and there was no reciprocal gain-loss within individual trials. card arrangement to balance the position effect. After The SGT contains four decks, each containing 5 card-turn- completing the computer game, subjects answered a ques- ing trials as a unit, for a total of 100 trials. The "bad" decks tionnaire that gauged their memories of task characteris- (defined by EV, as in IGT) in the SGT have worse EV ($ - tics and their deck preferences. 250) and better gain-loss frequency (4 gains and 1 loss), Page 3 of 10 (page number not for citation purposes) Behavioral and Brain Functions 2008, 4:13 http://www.behavioralandbrainfunctions.com/content/4/1/13 Table 2: A comparison of gamble structures between the Iowa Gambling Task and Soochow Gambling Task Iowa Gambling Task Gain-Loss Frequency EV (per 10 cards) Choice Prediction Based on Choice Prediction Based (per 10 cards) Gain-Loss Frequency on EV A (Bad) 5 G 5 L -$250 B (Bad) 9 G 1 L -$250 B C (Good) 6.25 G 1.25 L 2.5 S +$250 C C D (Good) 9 G 1 L +$250 D D Soochow Gambling (per 5 cards) (per 5 cards) Task * A (Bad) 4 G 1 L -$250 A B (Bad) 4 G 1 L -$250 B C (Good) 1 G 4 L +$250 C D (Good) 1 G 4 L +$250 D (G: gain, L: loss, S: standoff). * The deck structure for the Soochow Gambling Task is as follows: A, four consecutive gains of $200 followed by a loss of $1,050; B, four consecutive gains of $100 followed by a loss of $650; C, four consecutive losses of $200 followed by a gain of $1,050; D, four consecutive losses of $100 followed by a gain of $650. The experiment was using token money denominated in New Taiwan Dollars. For ease of comparison, the amount of money was nearly equated to U.S. currency in IGT. whereas the "good" decks have better EV ($ +250) and will make lose more than others. You can win if you stay away worse gain-loss frequency (1 gain and 4 losses). Thus, in from the worst decks ...." (Bechara et al., 1999, p. 5474). the SGT, the operation of gain-loss frequency and EV var- iables will predict different choice patterns. If decision- Results making is guided by EV, subjects will prefer the good Experimental results showed that subjects preferred the bad decks (A and B) to the good decks (C and D) (Figure decks (C and D) over the bad decks (A and B). Conversely, when participant choice behavior is controlled by imme- 1). This finding supports the effect of prediction-based diate gain-loss, the perceived gain will keep subjects on gain-loss frequency and is contrary to that of EV. the same deck. The deck with the highest number of gains will increase the probability of being chosen. Table 2 sum- Moreover, the average number of times subjects chose bad marizes a comparison of the task structures and predic- decks was higher than that for good decks throughout the tions between the IGT and the SGT. entire experiment (Figure 2). Significant interaction was found between gain-loss frequency and blocks. However, The SGT adopted an experimental procedure similar to no single main effect of blocks was observed. Subjects that used in the IGT to assure that subjects perform the gradually adjusted their selection pattern to the chance task under uncertain conditions. The probability struc- level of four choices. The absence of a crossover or lack of tures of gain-loss frequency were unknown to subjects in learning curve indicates that gain-loss frequency was both tasks (IGT and SGT). Subjects were also ignorant of dominant over EV when subjects were choosing decks. the time limitation in the experiment. Both tasks then are uncertain in the traditional sense. Furthermore, this investigation also analyzes the pattern of continuing choices after the subject chose AB or CD This study adopted the subject instructions employed by (Figure 3). The experimental results showed that subjects Bechara et al. [36]. The key points of the instructions are tend to remain on the same type of decks while choosing as follows: A or B. Nevertheless, no significant differences are found for the continuing choices between the same or the differ- "The goal of the game is to win as much money as possible and, ent types of decks while they were choosing C or D. if you find yourself unable to win, make sure you avoid losing money as much as possible. I won't tell you for how long the Analysis of the questionnaire results indicated that sub- game will continue. You must keep on playing until the compu- jects correctly recalled which decks they chose that had the ter stops. It is important to know that the colors of the cards are highest gains (Figure 4A); but subjects have a chance-level irrelevant in this game. The computer does not make you lose recollection on which decks they chose that had the high- money at random. However, there is no way for you to figure est losses (Figure 4B). Conversely, subjects incorrectly out when the computer will make you lose. All I can say is that recalled which decks were in use when they won the larg- you may find yourself losing money on all decks, but some decks est overall amount of money (Figure 4C); but subjects Page 4 of 10 (page number not for citation purposes) Behavioral and Brain Functions 2008, 4:13 http://www.behavioralandbrainfunctions.com/content/4/1/13 Mean Figure 2 number of cards chosen in blocks Mean number of cards chosen in blocks. 100 card selec- tion trials were grouped into five blocks, each comprising 20 trials. The three factor (repeated measurement) ANOVA (Gain-loss frequencies (gain vs. loss) × Values (± 200 vs. ± 100) × Blocks (1 to 5)) indicated a significant main effect for gain-loss frequencies (F (1, 47) = 29.44, p < .01) and values (F Mean n Figure 1 umber of cards chosen (1, 47) = 9.02, p < .01), but not for blocks (F (1, 47) = 0.00, p Mean number of cards chosen. The results of two fac- = 1.00). Furthermore, significant interactions existed tors (repeated measurement) ANOVA (Gain-loss frequen- between gain-loss frequencies and blocks (F (4, 44) = 3.03, p cies (gain vs. loss) × Values (± 200 vs. ± 100)) showed a < .05) as well as three factors (F (4, 44) = 5.19, p < .01); but significant effect of gain-loss frequency. Subjects selected non-significant interactions existed between gain-loss fre- more cards from decks A and B than decks C and D (F (1, quencies and values (F (1, 47) = 1.90, p = .18); values and 47) = 26.41, p < .01). The value effect with of the paired t- blocks (F (4, 44) = 0.99, p = .43). These results indicate a test is significant (t (47) = 2.60, p < .05) under high-frequency clear preference for the pooled decks A and B ("bad" decks) gain (+200 vs. +100), but not significant (t (47) = 0.72, p = over the pooled decks C and D ("good" decks) from the .48) under high-frequency loss (-200 vs. -100). None of the beginning. Subjects seem to be guided by gain-loss frequen- interaction effects are statistically significant (F (1, 47) = 1.88, cies and appear sensitive to the gain-loss structure gradually. p = .18). No cross-over or significant learning curve exists for the high-frequency gain (A, B) and high-frequency loss (C, D) decks under this condition (100 trials) in the Soochow Gam- have a chance-level to recall which decks were chosen bling Task. when they lost the largest overall amount of money (Fig- ure 4D). Preference patterns of conscious recollection were consistent with the choice pattern due to the effect of studies [45-49,52], therefore, the gain-loss frequency can gain-loss frequency (Figures 4E and 4F). successfully serve as a predictor for choice behavior under these uncertain situations. Discussion Subjects may apply an implicit strategy to cope with the The effect of gain-loss frequency is not an isolated finding uncertain game, therefore they favored high-frequency in similar settings. A reexamination of the bad decks (A gains over high-frequency losses in the experiment. This and B) in the original IGT [8] indicated that bad deck B (9 "gain-stay, lose-randomize" strategy (Figure 3) [42] has gains and 1 loss) was also chosen more frequently than been observed in human and animal appetitive and deck A (5 gains and 5 losses). Other studies obtained sim- avoidance experiments in which human or animal ilar findings but did not explore them further [18- encounter reward or punishment [42-48]. These pioneer 28,33,34]. Furthermore, some research groups even behavior studies with the concurrent schedules of rein- showed that normal subjects chose the disadvantageous forcement have displayed the frequency effect for choice deck B more frequently than the advantageous deck C or pattern [45-49]. Additionally, these concepts have also D [18-28]. Dunn et al. [14] sampled 38 IGT related studies been applied to examine the behavioral model of neu- to demonstrate that only five studies [18,19,23,25,53] uti- ropsychological deficit [50,51]. lized the "four-deck format" to display their findings (i.e., the number of card turnings for each deck over a total of The SGT possessed the variant frequency, magnitude and 100 trials being shown separately). These studies all dem- delay of reinforcement/punishment depending on each onstrated that deck B was chosen more frequently than subject's choice. Most part of these set-ups may be similar deck A. It is worth noting that four out of five studies to the consideration of the traditional behavior-analysis Page 5 of 10 (page number not for citation purposes) Behavioral and Brain Functions 2008, 4:13 http://www.behavioralandbrainfunctions.com/content/4/1/13 In the IGT, normal subjects shifted their choices to good decks during the latter part of the game, such that a learn- ing curve was evident [7-9,36,37,39]. We propose that this shift is not due to better EV but rather an effect resulting from more frequent gains than losses. The Somatic Marker Hypothesis stresses that somatic markers (or peripheral feedback) predispose normal subjects to behave in accordance with perceived future consequences over the long run. This study demonstrated that even subjects with intact somatic markers cannot behave accordingly to a search for EV in the SGT. Immediate reinforcement will override EV in the SGT. Decision makers may have been guided by immediate gain and, as such, their behavioral results are consistent with the prediction of gain-loss fre- quency. Based on the present observation, three possible interpretations exist for SMH: 1) somatic marker system may guide decision making behavior via rough-estima- tion processing (gain-loss frequency), not a precise calcu- Mean p Figure 3 robability of shift and stay in the continuing choice lation (EV: probability × value); 2) somatic marker system Mean probability of shift and stay in the continuing may only contribute to generating subjective feelings choice. After choosing the (bad) decks A or B, 65% of par- (consciousness), and may not be immediately related to ticipants remain on these two decks, with only 35% shifting decision guidance; 3) the operation of the somatic marker to the (good) decks C and D (t (47) = 8.60, p < .01). On the may be involved in gathering the long-term memory, but other hand, when subjects choose decks C and D, the proba- may not globally direct choice behavior in situations of bilities of them shifting or staying in their next selection was a high uncertainty. roughly 50/50 probability of selecting deck A and B versus C and D (t (47) = 1.71, p = .09). The Somatic Marker Hypothesis also posits that somatic markers guide advantageous behavior in a non-conscious manner. However, questionnaire results suggest further [18,19,23,25,53] also demonstrated that deck B was cho- that subjects can have a clear knowledge of gain and loss sen more than deck C or D. frequency by the end of the game. This perception eventu- ally determines their choice patterns. A similar finding Furthermore, in Peters and Slovic study [54], the modified was obtained by Maia and McClelland [34], demonstrat- IGT study also demonstrated that deck B ($ -250) and D ing a possible "conscious" knowledge of EV in the IGT. ($ +250) possessed the inversed expected values, but with nearly equal attraction to subjects. This may imply that In the SGT, normal subjects were stuck with the influence the expected value does not guide decision makers to of gain-loss frequency without shifting to EV throughout approach the beneficial choice in these dynamic games. the entire session. If normal subjects cannot resist the On the other hand, in the modified IGT, deck C contained influence of gain-loss frequency, ventromedial prefrontal high-frequency gain (8 out of 10 trials) over deck D (5 out patients would have increased vulnerability to the effect of of 10 trials), but with nearly equal expected value ($ +300 immediate gains and losses. This prediction seems to be in for C vs. $ +250 for D). However, subjects preferred deck line with disinhibition theory that suggests that socially C rather than deck D significantly. The gain-loss frequency dysfunctional patients with prefrontal damage have diffi- seems to be more reasonable than EV in explaining these culty avoiding a punishment-associated stimulus when observations. Ahn et al. [55] confirmed the present find- that stimulus was previously associated with a reward ing by comparing the decision learning models for IGT [13,15,16,56]. If this is true, then both normal subjects and SGT respectively. and ventromedial prefrontal patients will not be respon- sive to the long-term dimension. Questionnaire data in this study also indicate a novel phe- nomenon, namely, the "money account illusion". Sub- Supposing this is the case, the facilitative effect of somatic jects ignored the EV dimension and miscounted the markers did not induce a learning effect or an advanta- money amount in terms of the strength of frequency, spe- geous shifting behavior in the long run in the SGT, as con- cifically, frequently receiving gains will leave an overall sistently proposed by IGT researchers. Careful analysis of impression of large accumulated monetary outcome for a experimental results obtained in this study identified a deck than when gains are infrequently received. subordinate phenomenon in that deck A was chosen Page 6 of 10 (page number not for citation purposes) Behavioral and Brain Functions 2008, 4:13 http://www.behavioralandbrainfunctions.com/content/4/1/13 Subject m Figure 4 emory assessment in the Soochow Gambling Task Subject memory assessment in the Soochow Gambling Task. Forty-eight subjects were required to report their behavior and preferences after completion of the game. (4A) Most of the sample (36 subjects) had vivid impressions for high- 2 2 frequency gain decks (A+B) (x (1) = 12.00, p < .01), (4B) but not high-frequency loss (C+D) (x (1) = 0.00, p = 1.00). (4C) Additionally, most subjects (31 subjects recognized decks A and B) possessed a clear (but wrong) image for the overall mone- 2 2 tary gain (x (1) = 4.08, p < .05); (4D) but a blurred image for overall monetary loss (x (1) = .08, p = .77). (4E) After completing the game, subjects erroneously equated the high-frequency gains as the overall advantage. Thirty one out of forty eight subjects indicated the favorable choice to decks A and B rather than C and D (x (1) = 4.08, p < .05). (4F) Most unfavorable choice they had memorized were decks C and D which possessed the high-frequency loss and positive expected value (x (1) = 8.33, p < .01). Page 7 of 10 (page number not for citation purposes) Behavioral and Brain Functions 2008, 4:13 http://www.behavioralandbrainfunctions.com/content/4/1/13 more often than deck B, despite both decks possessing the Competing interests same gain-loss frequency. The immediate value of gain The authors certify that the information listed above is and loss can also alter slightly decision-makers' choices complete to the best of our original research. The authors and is worthy of further exploration. declare that they have no competing interests. EV and gain-loss frequency were seriously confounded in Authors' contributions the administration of the IGT. Selection of good decks by This design of Soochow Gambling Task was constructed normal subjects cannot only be attributed to the effect of by YC and CH, both have made equal contributions to EV, but must also be explained in terms of gain-loss fre- thought, data interpretation, drafting the key concept, and quency. The Somatic Marker Hypothesis was further rein- revising it critically. JT provided several times of critical forced by adopting the principal findings from the IGT review to organize the whole picture for this work and experiment as important supporting data, i.e., somatic revised this article significantly. PL worked on the compu- markers predispose normal subjects to search for EV in the terization of task, consulting of data analysis. SY and JC long run. This study explored separately the relative con- participated in some discussion and provided some valu- tribution of EV and gain-loss frequency in the SGT, a mod- able viewpoints to improve the meaning of data and read- ified gambling task. Experimental results indicated that ability. All authors gave final approval of the version to be immediate reinforcement overrides the effect of EV. Crone published. et al. [57] also identified a local preference for gain-loss frequency; however, EV dominated gain-loss frequency in Acknowledgements The authors would like to thank the National Science Council of the their modified IGT. In contrast to the predictions based on Republic of China, Taiwan, for financially supporting this research under Somatic Marker Hypothesis, experimental data in this Contract No. NSC 94-2413-H031-002. Professors Keng-Chen Liang and study indicate that normal subjects were primarily guided Daisy Hung are appreciated for their valuable courses for introduction of by the effect of gain-loss frequency rather than EV. This Somatic Marker Hypothesis. Dr. Ian Tomb and Professor Marc Hauser are finding, although differing from the simple explanation also appreciated for kindly providing the original instruction of IGT. Thanks provided by the Somatic Marker Hypothesis, is consistent also extend to the editors and reviewers of several journals over the past with the view of behavioral [1-5] and affective [5,20,58- 4 years for their valuable comments. 62] decision literature, indicating that normal individuals are often short-sighted when making decisions in stock References 1. Hastie R, Dawes RM: Rational Decision in an uncertainty world: market or real life [41,63,64]. The psychology of judgment and decision making. London: Sage Publications; 2001. Conclusion 2. Kahneman D: Maps of bounded rationality: Psychology for behavioral economics. Amer Econ Rev 2003, 93:1449-1475. A serious confounding effect in IGT was demonstrated by 3. Kahneman D, Tversky A: Prospect theory: An analysis of deci- the "prominent deck B phenomenon". This recently-dis- sion under risk. Econometrica 1979, 47:263-291. 4. Tversky A, Kahneman D: The framing of decisions and the psy- covered phenomenon may imply that gain-loss frequency chology of choice. Science 1981, 211:453-458. rather than EV dominates the choice behavior of normal 5. Panksepp J: Damasio's Error. Consciousness and Emotion 2003, decision makers under uncertainty. In the original IGT, 4:111-134. 6. Lichtenstein S, Slovic P, Zink DJ: Effect of instruction in expected these two factors were seriously confounded, and thus this value on optimality of gambling decisions. J Exp Psychol Gen study proposed a modified task SGT to clarify the relative 1969, 79:236-240. contribution of these two factors. The experimental results 7. Damasio AR: Descartes' error: emotion, reason, and the human brain. New York: G. P. Putnam's Sons; 1994. indicated that subject selection patterns were mostly pre- 8. Bechara A, Damasio AR, Damasio H, Anderson SW: Insensitivity to dicted based on gain-loss frequency. Immediate gain may future consequences following damage to human prefrontal cortex. Cognition 1994, 50:7-15. increase the probability of continuing to stick on the cho- 9. Bechara A, Damasio H, Tranel D, Damasio AR: Deciding advanta- sen deck, which is consistent with most behavioral-analy- geously before knowing the advantageous strategy. Science sis and decision literatures. The observations of this study 1997, 275:1293-1295. 10. Leland J, Grafman J: Experimental tests of Somatic Marker indicate that EV is not predictive for choice behaviors as hypothesis. Games Econ Behav 2005, 52:386-409. suggested by IGT. It seems that immediate gain eventually 11. Tomb I, Hauser M, Deldin P, Caramazza A: Do somatic markers mediate decisions on the gambling task? Nat Neurosci 2002, resulted in a long-term loss. The basic assumption of IGT 5:1103-1104. is thus no longer plausible to assure the foresighted deci- 12. Suzuki A, Hirota A, Takasawa N, Shigemasu K: Application of the sion makers in the Iowa Gambling Task. somatic marker hypothesis to individual differences in deci- sion making. Biol Psychol 2003, 65(1):81-88. 13. Rolls ET: Emotion Explained. New York: Oxford University Press; Abbreviations EV: Expected Value; IGT: Iowa Gambling Task; SGT: Soo- 14. Dunn BD, Dalgleish T, Lawrence AD: The somatic marker hypothesis: A critical evaluation. Neurosci Biobehav Rev 2006, chow Gambling Task; SMH: Somatic Marker Hypothesis 30(2):239-271. 15. Rolls ET: The brain and emotion. New York: Oxford University Press; 1999. Page 8 of 10 (page number not for citation purposes) Behavioral and Brain Functions 2008, 4:13 http://www.behavioralandbrainfunctions.com/content/4/1/13 16. Kringelbach ML, Rolls ET: The functional neuroanatomy of the 37. Bechara A, Tranel D, Damasio H, Damasio AR: Failure to respond human orbitofrontal cortex: evidence from neuroimaging autonomically to anticipated future outcomes following and neuropsychology. Prog Neurobiol 2004, 72:341-372. damage to prefrontal cortex. Cereb Cortex 1996, 6:215-225. 17. Lin CH, Chiu YC, Huang JT: Growing evidence of prominent 38. Bechara A, Damasio H, Tranel D, Anderson SW: Dissociation Of deck B phenomenon in Iowa Gambling Task. Society of Neur- working memory from decision making within the human oeconomics 5th Annual Meeting: 2007; Boston, MA 2007. prefrontal cortex. J Neurosci 1998, 18(1):428-437. 18. Bark R, Dieckmann S, Bogerts B, Northoff G: Deficit in decision 39. Bechara A, Tranel D, Damasio H, Damasio AR: Characterization making in catatonic schizophrenia: an exploratory study. Psy- of the decision-making deficit of patients with ventromedial chiatry Res 2005, 134(2):131-141. prefrontal cortex lesions. Brain 2000, 123:2189-2202. 19. Wilder KE, Weinberger DR, Goldberg TE: Operant conditioning 40. Chiu YC, Lin CH, Huang JT, Lin S, Lee PL, Hsieh JC: Immediate gain and the orbitofrontal cortex in schizophrenic patients: unex- is long-term loss: Are there foresighted decision makers in pected evidence for intact functioning. Schizophr Res 1998, Iowa gambling task? Society for Neuroeconomics 3rd Annual Meeting: 30(2):169-174. September 15–18 2005; Kiawah Island, South Carolina, U.S.A. 2005. 20. O'Carroll RE, Papps BP: Decision making in humans: the effect 41. Chiu YC, Lin CH, Lin S, Huang JT: Reexamining the Effect of of manipulating the central noradrenergic system. J Neurol Long-term Outcome and Gain-loss Frequency: From Uncer- Neurosurg Psychiatry 2003, 74(3):376-378. tainty to Certainty. Society for Neuroeconomics 4th Annual Meeting: 21. Crone EA, Somsen RJM, van Beek B, van Der Molen MW: Heart 2006; Park City, Utah, U.S.A. 2006. rate and skin conductance analysis of antecendents and con- 42. Mitropoulos A: An experiment on the value of structural infor- sequences of decision making. Psychophysiology 2004, mation in a 2 × 2 repeated game. Econ Letters 2003, 78:27-32. 41:531-540. 43. Colman AM: Oxford Dictionary of Psychology. 1st edition. New 22. Overman WH, Frassrand K, Ansel S, Trawalter S, Bies B, Redmond York: Oxford University Press; 2003. A: Performance on the IOWA card task by adolescents and 44. Herrnstein RJ: Formal properties of the matching law. J Exp adults. Neuropsychologia 2004, 42(13):1838-1851. Anal Behav 1974, 21:159-164. 23. Ritter LM, Meador-Woodruff JH, Dalack GW: Neurocognitive 45. Rachlin H: Introduction to modern behaviorism. 3rd edition. measures of prefrontal cortical dysfunction in schizophrenia. New York: W. H. Freeman and Company; 1991. Schizophr Res 2004, 68(1):65-73. 46. Schwartz B, Wasserman EA, Robbins SJ: Psychology of learning 24. Rodriguez-Sanchez JM, Crespo-Facorro B, Perez-Iglesias R, Gonzalez- and Behavior. 5th edition. New York: W. W. Norton & Company; Blanch C, Alvarez-Jimenez M, Llorca J, Vazquez-Barquero JL: Pre- 2002. frontal cognitive functions in stabilized first-episode patients 47. Catania AC, Sagvolden T: Preference for free choice over forced with schizophrenia spectrum disorders: a dissociation choice in pigeons. J Exp Anal Behav 1980, 34:77-86. between dorsolateral and orbitofrontal functioning. Schizophr 48. Catania AC: Learning. 4th edition. NJ: Prentice Hall: Upper Saddle Res 2005, 77(2–3):279-288. River; 1998. 25. Toplak ME, Jain U, Tannock R: Executive and motivational proc- 49. Domjan M: The priciples of learning and behavior. fourth edi- esses in adolescents with Attention-Deficit-Hyperactivity tion. Belmont, CA: Wadsworth/Thompson; 2002. Disorder (ADHD). Behav Brain Funct 2005, 1(1):8. 50. Sagvolden T, Aase H, Zeiner P, Berger D: Altered reinforcement 26. Fernie G, Tunney RJ: Some decks are better than others: the mechanisms in attention-deficit/hyperactivity disorder. effect of reinforcer type and task instructions on learning in Behav Brain Res 1998, 94(1):61-71. the Iowa Gambling Task. Brain Cogn 2006, 60(1):94-102. 51. Johansen EB, Killeen PR, Sagvolden T: Behavioral variability, elim- 27. Caroselli JS, Hiscock M, Scheibel RS, Ingram F: The simulated gam- ination of responses, and delay-of-reinforcement gradients bling paradigm applied to young adults: an examination of in SHR and WKY rats. Behav Brain Funct 2007, 3:60. university students' performance. Appl Neuropsychol 2006, 52. Ferster CB, Skinner BF: Schedules of reinforcement. MA: Pren- 13(4):203-212. tice-Hall, Inc; 1957. 28. Martino DJ, Bucay D, Butman JT, Allegri RF: Neuropsychological 53. Shurman B, Horan WP, Nuechterlein KH: Schizophrenia patients frontal impairments and negative symptoms in schizophre- demonstrate a distinctive pattern of decision-making nia. Psychiatry Res 2007. impairment on the Iowa Gambling Task. Schizophr Res 2005, 29. Lin CH, Chiu YC, Lee PL, Hsieh JC: Is deck B a disadvantageous 72(2–3):215-224. deck in the Iowa Gambling Task? Behav Brain Funct 2007, 3:16. 54. Peters E, Slovic P: The springs of action: Affective and analyti- 30. Sevy S, Burdick KE, Visweswaraiah H, Abdelmessih S, Lukin M, cal Information processing in choice. Pers Soc Psychol Bull 2000, Yechiam E, Bechara A: Iowa gambling task in schizophrenia: a 26:1465-1475. review and new data in patients with schizophrenia and co- 55. Ahn WY, Busemeyer JR, Wagenmakers EJ, Stout JC: Comparison of occurring cannabis use disorders. Schizophr Res 2007, 92(1– decision learning models using the generalization criterion 3):74-84. method. Cogn Sci 2008 in press. Accepted pending revisions 31. Johnson CA, Xiao L, Palmer P, Sun P, Wang Q, Wei Y, Jia Y, Grenard 56. Rolls ET, Hornak J, Wade D, McGrath J: Emotion-related learning JL, Stacy AW, Bechara A: Affective decision-making deficits, in patients with social and emotional changes associated linked to a dysfunctional ventromedial prefrontal cortex, with frontal lobe damage. J Neurol Neurosurg Psychiatry 1994, revealed in 10th grade Chinese adolescent binge drinkers. 57:1518-1524. Neuropsychologia 2008, 46(2):714-726. 57. Crone EA, van der Molen MW: Developmental changes in real 32. Chiu YC, Lin CH: Is deck C an advantageous deck in the Iowa life decision making: performance on a gambling task previ- Gambling Task? Behav Brain Funct 2007, 3(37):. ously shown to depend on the ventromedial prefrontal cor- 33. Lin CH, Chiu YC, Lee PL, Hsieh JC: The preferences of decision- tex. Dev Neuropsychol 2004, 25(3):251-279. making in IOWA Gambling Task: The testing of frequency 58. Berridge K: Irrational Pursuits: Hyper-Incentives from a Vis- effect and long-term outcomes in Somatic Marker Hypothe- ceral Brain. In The psychology of economics decisions: Volume I: Ration- sis. 2nd Conference of NeuroEconomics: 2004; Münster, Germany 2004. ality and Well-Being Edited by: Brocas I, Carrillo J. New York: Oxford 34. Maia TV, McClelland JL: A reexamination of the evidence for the University Press; 2003:17-40. somatic marker hypothesis: what participants really know in 59. Baumeister R: The psychology of irrationality: Why people the Iowa gambling task. Proc Natl Acad Sci USA 2004, make foolish, self-defeating choices. In The psychology of econom- 101(45):16075-16080. ics decisions: Volume I: Rationality and Well-Being Edited by: Brocas I, 35. Bowman CH, Evans CEY, Turnbull OH: Artificial time constraints Carrillo J. New York: Oxford University Press; 2003. on the Iowa Gambling Task: The effects on behavioral per- 60. Heims HC, Critchley HD, Dolan R, Mathias CJ, Cipolotti L: Social formance and subjective experience. Brain Cogn 2005, and motivational functioning is not critically dependent on 57:21-25. feedback of autonomic responses: neuropsychological evi- 36. Bechara A, Damasio H, Damasio AR, Lee GP: Different contribu- dence from patients with pure autonomic failure. Neuropsych- tions of the human amygdala and ventromedial prefrontal ologia 2004, 42(14):1979-1988. cortex to decision-making. J Neurosci 1999, 19:5473-5481. 61. North NT, O'Carroll RE: Decision making in patients with spi- nal cord damage: afferent feedback and the somatic marker hypothesis. Neuropsychologia 2001, 39(5):521-524. Page 9 of 10 (page number not for citation purposes) Behavioral and Brain Functions 2008, 4:13 http://www.behavioralandbrainfunctions.com/content/4/1/13 62. Ross D, Spurrett D, Vuchinich R: The behavioral economics and neuroeconomics disordered gambling: A policy-focused sur- vey of research. A report prepared for the South African Responsible Gambling Foundation. Cape Town 2006:1-255. 63. Peterson R: Inside the Investor's Brain: The Power of Mind Over Money. SA: John Wiley & Sons; 2007. 64. Lin CH, Chiu YC, Huang JT: Is decision-maker sensitive to expected value in the dynamic-uncertain gambles? 5th Annual Meeting of Society for Neuroeconomics: 2007; Boston, MA 2007. Publish with Bio Med Central and every scientist can read your work free of charge "BioMed Central will be the most significant development for disseminating the results of biomedical researc h in our lifetime." Sir Paul Nurse, Cancer Research UK Your research papers will be: available free of charge to the entire biomedical community peer reviewed and published immediately upon acceptance cited in PubMed and archived on PubMed Central yours — you keep the copyright BioMedcentral Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp Page 10 of 10 (page number not for citation purposes) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Behavioral and Brain Functions Springer Journals

Immediate gain is long-term loss: Are there foresighted decision makers in the Iowa Gambling Task?

Loading next page...
 
/lp/springer-journals/immediate-gain-is-long-term-loss-are-there-foresighted-decision-makers-gUpuTtWehS

References (62)

Publisher
Springer Journals
Copyright
Copyright © 2008 by Chiu et al; licensee BioMed Central Ltd.
Subject
Biomedicine; Neurosciences; Neurology; Behavioral Therapy; Psychiatry
eISSN
1744-9081
DOI
10.1186/1744-9081-4-13
pmid
18353176
Publisher site
See Article on Publisher Site

Abstract

Background: The Somatic Marker Hypothesis suggests that normal subjects are "foreseeable" and ventromedial prefrontal patients are "myopic" in making decisions, as the behavior shown in the Iowa Gambling Task. The present study questions previous findings because of the existing confounding between long-term outcome (expected value, EV) and gain-loss frequency variables in the Iowa Gambling Task (IGT). A newly and symmetrically designed gamble, namely the Soochow Gambling Task (SGT), with a high-contrast EV between bad (A, B) and good (C, D) decks, is conducted to clarify the issue about IGT confounding. Based on the prediction of EV (a basic assumption of IGT), participants should prefer to choose good decks C and D rather than bad decks A and B in SGT. In contrast, according to the prediction of gain-loss frequency, subjects should prefer the decks A and B because they possessed relatively the high-frequency gain. Methods: The present experiment was performed by 48 participants (24 males and 24 females). Most subjects are college students recruited from different schools. Each subject played the computer version SGT first and completed a questionnaire for identifying their final preference. The IGT experimental procedure was mostly followed to assure a similar condition of decision uncertainty. Results: The SGT experiment demonstrated that the prediction of gain-loss frequency is confirmed. Most subjects preferred to choose the bad decks A and B than good decks C and D. The learning curve and questionnaire data indicate that subjects can not "hunch" the EV throughout the game. Further analysis of the effect of previous choice demonstrated that immediate gain increases the probability to stay at the same deck. Conclusion: SGT provides a balanced structure to clarify the confounding inside IGT and demonstrates that gain-loss frequency rather than EV guides decision makers in these high-ambiguity gambles. Additionally, the choice behavior is mostly following the "gain-stay, lose-randomize" strategy to cope with the uncertain situation. As demonstrated in SGT, immediate gain can bring about a long-term loss under uncertainty. This empirical result may explain some shortsighted behaviors in real life. Page 1 of 10 (page number not for citation purposes) Behavioral and Brain Functions 2008, 4:13 http://www.behavioralandbrainfunctions.com/content/4/1/13 "... the brains of the normal subjects were gradually learning to Background Studies in behavioral decision-making and affective neu- predict a bad outcome, and were signaling the relative badness roscience have found that typical decision makers are fre- of the particular deck before the actual card-turning." quently "myopic" [1-5] to long-term outcome (expected (Damasio, 1994, p 220). value, EV) [6]. Conversely, Damasio [7] and Bechara et al. [8,9] proposed the Somatic Marker Hypothesis and con- In short, the Somatic Marker Hypothesis suggests that ducted the Iowa Gambling Task (IGT) to test whether ven- somatic markers, which are processed implicitly, can facil- tromedial prefrontal patients are shortsighted in terms of itate decision makers in making advantageous decisions the future and long-term outcome, and whether typical [9] and guiding explicit decision making [33,34]. decision makers can predict or foresee the future. For comparison, the notion of shortsighted vs. foresighted is The most compelling empirical evidence for supporting adopted in this study. A marked difference exists between the Somatic Marker Hypothesis is found in the IGT [8]. shortsightedness and foresight. The IGT was the central Over the past decade, the IGT has been widely employed test for a verification of the Somatic Marker Hypothesis. as a neuropsychological research instrument for investi- Some studies have attempted to replicate or modify the gating affective and executive function. At least 100 scien- protocol used in the IGT [10-12]; whereas others have tific studies have utilized the IGT to investigate a diverse proposed that the hypothesis was theoretically inade- set of neurological and psychiatric populations [35]. quate [1,5,13-16]. Lin et al. [17] pointed out that there are increasing number of studies [18-28] to demonstrate a In the IGT, card decks A and B were designated "bad" contradictory phenomenon, namely, the "prominent decks with low EV ($ -250), and C and D were "good" deck B phenomenon" [29]. The phenomenon showed decks with high EV ($ +250) in average 10 trials. Subjects that normal decision makers can not prevent their prefer- and ventromedial prefrontal patients selected "bad" decks ence to "bad" (EV) deck B in the standard version of IGT during the first 30 card-turning trials (out of 100) [7,36]; (or, due to the effect of "gain-loss frequency"). Recently, however, normal subjects gradually shifted to "good" Bechara (Sevy et al.) [30,31] also revealed a "prominent decks and avoided the "bad" decks [8,9,36-39]. However, deck B phenomenon". In their study, normal subjects pre- a careful examination of the IGT reveals a critical con- ferred the bad deck B rather than the other three decks. founding between EV and gain-loss frequency. On aver- Furthermore, the chosen number of deck B (31) in Sevy et age, for each 10-card unit, deck A contains 5 gains and 5 al. study [30] is almost a double of Bechara et al. data in losses, and deck B contains 9 gains and 1 loss. On the 1994 (about 17) [8]. On the other hand, Lin et al. [29] other hand, deck C contains 6.25 gains, 2.5 standoffs and analyzed the existing experimental results adopting sim- 1.25 losses, and D contains 9 gains and 1 loss. Altogether ple version of IGT and suggest that the "prominent deck B the bad decks (A and B) contain 14 gains and 6 losses, phenomenon" may be due to a confounding from gain- whereas good decks (C and D) contain 15.25 gains, 2.5 loss frequency. Additionally, Chiu and Lin [32] utilize a standoffs, and only 2.25 losses. Both good and bad decks modified version of IGT to demonstrate that subjects' have a similar number of gains, whereas the good decks preference to deck C is also due to gain-loss frequency, not have significantly fewer losses (see Tables 1 and 2). There- EV. A fundamental structural flaw exists in a failure of fore, it is not clear whether subjects' choices of good decks orthogonal separation between EV and gain-loss fre- in the IGT were driven by improved EV or gain-loss fre- quency (or, frequency of punishments and rewards). This quency. This study attempts to generate a symmetrical and limitation has not been properly evaluated in IGT litera- fair experiment to demonstrate the relative guiding power ture. The study tries to explore the implications of this of gain-loss frequency and EV under uncertainty, specifi- flaw. cally to identify which factor most comprehensively dom- inates normal subject preferences. To differentiate Damasio's [7] Somatic Marker Hypothesis proposes that between the relative contributions of EV and gain-loss fre- normal decision-making is often assisted by somatic quency, this study applies a new task, namely, the Soo- markers. Ventromedial prefrontal patients are influenced chow Gambling Task (SGT) [40,41] (see Table 1). largely by immediate reinforcement and are insensitive to, or cannot see, future consequences due to a lack of past Method affective experiences. Normal subjects with intact somatic A two-phase study was designed to explore learning proc- markers benefit from repeated exposure to punishments ess and final preferences during decision making. Four and rewards when performing tasks and are cognizant of card decks, randomized into 24 arrangements (e.g., future outcome. Somatic states serve as a neural expres- ABCD, BCDA, CDAB, etc.), were presented on a computer sion biasing the brain process when evaluating the bad- screen. Forty-eight college students and adults (24 males, ness or goodness of each decision. According to Damasio 24 females; age: 18–30 years, mean age: 20.71 years) par- [7]: ticipated in this study. Each subject undertook one set of Page 2 of 10 (page number not for citation purposes) Behavioral and Brain Functions 2008, 4:13 http://www.behavioralandbrainfunctions.com/content/4/1/13 Table 1: The immediate net value of each trial and gain-loss structure in the original IGT and SGT IGT Serial Numbers AB C D SGT AB C D 1 100 100 50 50 1 200 100 -200 -100 2 100 100 50 50 2 200 100 -200 -100 3 -50 100 0 50 3 200 100 -200 -100 4 100 100 50 50 4 200 100 -200 -100 5 -200 100 0 50 5 -1050 -650 1050 650 6 100 100 50 50 6 200 100 -200 -100 7 -100 100 0 50 7 200 100 -200 -100 8 100 100 50 50 8 200 100 -200 -100 9 -150 -1150 050 9 200 100 -200 -100 10 -250 100 0 -200 10 -1050 -650 1050 650 11 100 100 50 50 11 200 100 -200 -100 12 -250 100 25 50 12 200 100 -200 -100 13 100 100 -25 50 13 200 100 -200 -100 14 -150 -1150 50 50 14 200 100 -200 -100 15 -100 100 50 50 15 -1050 -650 1050 650 16 100 100 50 50 16 200 100 -200 -100 17 -200 100 25 50 17 200 100 -200 -100 18 -50 100 -25 50 18 200 100 -200 -100 19 100 100 50 50 19 200 100 -200 -100 20 100 100 0 -200 20 -1050 -650 1050 650 21 100 -1150 50 50 21 200 100 -200 -100 22 -200 100 50 50 22 200 100 -200 -100 23 100 100 50 50 23 200 100 -200 -100 24 -250 100 0 50 24 200 100 -200 -100 25 100 100 25 50 25 -1050 -650 1050 650 26 -100 100 0 50 26 200 100 -200 -100 27 -150 100 50 50 27 200 100 -200 -100 28 -50 100 50 50 28 200 100 -200 -100 29 100 100 -25 -200 29 200 100 -200 -100 30 100 100 0 50 30 -1050 -650 1050 650 31 -250 100 50 50 31 200 100 -200 -100 32 -100 -1150 50 50 32 200 100 -200 -100 33 -150 100 50 50 33 200 100 -200 -100 34 100 100 25 50 34 200 100 -200 -100 35 100 100 25 -200 35 -1050 -650 1050 650 36 100 100 50 50 36 200 100 -200 -100 37 -50 100 -25 50 37 200 100 -200 -100 38 -200 100 50 50 38 200 100 -200 -100 39 100 100 0 50 39 200 100 -200 -100 40 100 100 -25 50 40 -1050 -650 1050 650 Final Outcomes -1000 ($) -1000 ($) +1000 ($) +1000 ($) Final Outcomes -2000 ($) -2000 ($) +2000 ($) +2000 ($) Note. In these gambling tasks, the internal gamble structure of IGT and SGT and the trial to end the game is unknown to the subjects. The subjects are asked to minimize monetary expenditure and maximize their winnings by choosing one card from the four decks in each trial. In the table, negative values are marked with a bold font size. (Left part: IGT) Deck A contains the relative high-frequency loss, while decks B, C, and D contain the high-frequency gain (net value within each trial). Decks A and B have negative net value (namely, the EV), -250($) over an average of ten trials; moreover, C and D have a positive net value of +250($) over an average of ten trials. (Right part: SGT) Five cumulative trials are repeated for each deck in the Soochow Gambling Task, decks A and B have high-frequency gain, while decks C and D exhibit a reversed gain-loss pattern. The task enlarges the difference between positive and negative EVs to make the difference more noticeable than in the Iowa Gambling Task. While playing this game, subjects only experienced a gain or a loss during each trial, and there was no reciprocal gain-loss within individual trials. card arrangement to balance the position effect. After The SGT contains four decks, each containing 5 card-turn- completing the computer game, subjects answered a ques- ing trials as a unit, for a total of 100 trials. The "bad" decks tionnaire that gauged their memories of task characteris- (defined by EV, as in IGT) in the SGT have worse EV ($ - tics and their deck preferences. 250) and better gain-loss frequency (4 gains and 1 loss), Page 3 of 10 (page number not for citation purposes) Behavioral and Brain Functions 2008, 4:13 http://www.behavioralandbrainfunctions.com/content/4/1/13 Table 2: A comparison of gamble structures between the Iowa Gambling Task and Soochow Gambling Task Iowa Gambling Task Gain-Loss Frequency EV (per 10 cards) Choice Prediction Based on Choice Prediction Based (per 10 cards) Gain-Loss Frequency on EV A (Bad) 5 G 5 L -$250 B (Bad) 9 G 1 L -$250 B C (Good) 6.25 G 1.25 L 2.5 S +$250 C C D (Good) 9 G 1 L +$250 D D Soochow Gambling (per 5 cards) (per 5 cards) Task * A (Bad) 4 G 1 L -$250 A B (Bad) 4 G 1 L -$250 B C (Good) 1 G 4 L +$250 C D (Good) 1 G 4 L +$250 D (G: gain, L: loss, S: standoff). * The deck structure for the Soochow Gambling Task is as follows: A, four consecutive gains of $200 followed by a loss of $1,050; B, four consecutive gains of $100 followed by a loss of $650; C, four consecutive losses of $200 followed by a gain of $1,050; D, four consecutive losses of $100 followed by a gain of $650. The experiment was using token money denominated in New Taiwan Dollars. For ease of comparison, the amount of money was nearly equated to U.S. currency in IGT. whereas the "good" decks have better EV ($ +250) and will make lose more than others. You can win if you stay away worse gain-loss frequency (1 gain and 4 losses). Thus, in from the worst decks ...." (Bechara et al., 1999, p. 5474). the SGT, the operation of gain-loss frequency and EV var- iables will predict different choice patterns. If decision- Results making is guided by EV, subjects will prefer the good Experimental results showed that subjects preferred the bad decks (A and B) to the good decks (C and D) (Figure decks (C and D) over the bad decks (A and B). Conversely, when participant choice behavior is controlled by imme- 1). This finding supports the effect of prediction-based diate gain-loss, the perceived gain will keep subjects on gain-loss frequency and is contrary to that of EV. the same deck. The deck with the highest number of gains will increase the probability of being chosen. Table 2 sum- Moreover, the average number of times subjects chose bad marizes a comparison of the task structures and predic- decks was higher than that for good decks throughout the tions between the IGT and the SGT. entire experiment (Figure 2). Significant interaction was found between gain-loss frequency and blocks. However, The SGT adopted an experimental procedure similar to no single main effect of blocks was observed. Subjects that used in the IGT to assure that subjects perform the gradually adjusted their selection pattern to the chance task under uncertain conditions. The probability struc- level of four choices. The absence of a crossover or lack of tures of gain-loss frequency were unknown to subjects in learning curve indicates that gain-loss frequency was both tasks (IGT and SGT). Subjects were also ignorant of dominant over EV when subjects were choosing decks. the time limitation in the experiment. Both tasks then are uncertain in the traditional sense. Furthermore, this investigation also analyzes the pattern of continuing choices after the subject chose AB or CD This study adopted the subject instructions employed by (Figure 3). The experimental results showed that subjects Bechara et al. [36]. The key points of the instructions are tend to remain on the same type of decks while choosing as follows: A or B. Nevertheless, no significant differences are found for the continuing choices between the same or the differ- "The goal of the game is to win as much money as possible and, ent types of decks while they were choosing C or D. if you find yourself unable to win, make sure you avoid losing money as much as possible. I won't tell you for how long the Analysis of the questionnaire results indicated that sub- game will continue. You must keep on playing until the compu- jects correctly recalled which decks they chose that had the ter stops. It is important to know that the colors of the cards are highest gains (Figure 4A); but subjects have a chance-level irrelevant in this game. The computer does not make you lose recollection on which decks they chose that had the high- money at random. However, there is no way for you to figure est losses (Figure 4B). Conversely, subjects incorrectly out when the computer will make you lose. All I can say is that recalled which decks were in use when they won the larg- you may find yourself losing money on all decks, but some decks est overall amount of money (Figure 4C); but subjects Page 4 of 10 (page number not for citation purposes) Behavioral and Brain Functions 2008, 4:13 http://www.behavioralandbrainfunctions.com/content/4/1/13 Mean Figure 2 number of cards chosen in blocks Mean number of cards chosen in blocks. 100 card selec- tion trials were grouped into five blocks, each comprising 20 trials. The three factor (repeated measurement) ANOVA (Gain-loss frequencies (gain vs. loss) × Values (± 200 vs. ± 100) × Blocks (1 to 5)) indicated a significant main effect for gain-loss frequencies (F (1, 47) = 29.44, p < .01) and values (F Mean n Figure 1 umber of cards chosen (1, 47) = 9.02, p < .01), but not for blocks (F (1, 47) = 0.00, p Mean number of cards chosen. The results of two fac- = 1.00). Furthermore, significant interactions existed tors (repeated measurement) ANOVA (Gain-loss frequen- between gain-loss frequencies and blocks (F (4, 44) = 3.03, p cies (gain vs. loss) × Values (± 200 vs. ± 100)) showed a < .05) as well as three factors (F (4, 44) = 5.19, p < .01); but significant effect of gain-loss frequency. Subjects selected non-significant interactions existed between gain-loss fre- more cards from decks A and B than decks C and D (F (1, quencies and values (F (1, 47) = 1.90, p = .18); values and 47) = 26.41, p < .01). The value effect with of the paired t- blocks (F (4, 44) = 0.99, p = .43). These results indicate a test is significant (t (47) = 2.60, p < .05) under high-frequency clear preference for the pooled decks A and B ("bad" decks) gain (+200 vs. +100), but not significant (t (47) = 0.72, p = over the pooled decks C and D ("good" decks) from the .48) under high-frequency loss (-200 vs. -100). None of the beginning. Subjects seem to be guided by gain-loss frequen- interaction effects are statistically significant (F (1, 47) = 1.88, cies and appear sensitive to the gain-loss structure gradually. p = .18). No cross-over or significant learning curve exists for the high-frequency gain (A, B) and high-frequency loss (C, D) decks under this condition (100 trials) in the Soochow Gam- have a chance-level to recall which decks were chosen bling Task. when they lost the largest overall amount of money (Fig- ure 4D). Preference patterns of conscious recollection were consistent with the choice pattern due to the effect of studies [45-49,52], therefore, the gain-loss frequency can gain-loss frequency (Figures 4E and 4F). successfully serve as a predictor for choice behavior under these uncertain situations. Discussion Subjects may apply an implicit strategy to cope with the The effect of gain-loss frequency is not an isolated finding uncertain game, therefore they favored high-frequency in similar settings. A reexamination of the bad decks (A gains over high-frequency losses in the experiment. This and B) in the original IGT [8] indicated that bad deck B (9 "gain-stay, lose-randomize" strategy (Figure 3) [42] has gains and 1 loss) was also chosen more frequently than been observed in human and animal appetitive and deck A (5 gains and 5 losses). Other studies obtained sim- avoidance experiments in which human or animal ilar findings but did not explore them further [18- encounter reward or punishment [42-48]. These pioneer 28,33,34]. Furthermore, some research groups even behavior studies with the concurrent schedules of rein- showed that normal subjects chose the disadvantageous forcement have displayed the frequency effect for choice deck B more frequently than the advantageous deck C or pattern [45-49]. Additionally, these concepts have also D [18-28]. Dunn et al. [14] sampled 38 IGT related studies been applied to examine the behavioral model of neu- to demonstrate that only five studies [18,19,23,25,53] uti- ropsychological deficit [50,51]. lized the "four-deck format" to display their findings (i.e., the number of card turnings for each deck over a total of The SGT possessed the variant frequency, magnitude and 100 trials being shown separately). These studies all dem- delay of reinforcement/punishment depending on each onstrated that deck B was chosen more frequently than subject's choice. Most part of these set-ups may be similar deck A. It is worth noting that four out of five studies to the consideration of the traditional behavior-analysis Page 5 of 10 (page number not for citation purposes) Behavioral and Brain Functions 2008, 4:13 http://www.behavioralandbrainfunctions.com/content/4/1/13 In the IGT, normal subjects shifted their choices to good decks during the latter part of the game, such that a learn- ing curve was evident [7-9,36,37,39]. We propose that this shift is not due to better EV but rather an effect resulting from more frequent gains than losses. The Somatic Marker Hypothesis stresses that somatic markers (or peripheral feedback) predispose normal subjects to behave in accordance with perceived future consequences over the long run. This study demonstrated that even subjects with intact somatic markers cannot behave accordingly to a search for EV in the SGT. Immediate reinforcement will override EV in the SGT. Decision makers may have been guided by immediate gain and, as such, their behavioral results are consistent with the prediction of gain-loss fre- quency. Based on the present observation, three possible interpretations exist for SMH: 1) somatic marker system may guide decision making behavior via rough-estima- tion processing (gain-loss frequency), not a precise calcu- Mean p Figure 3 robability of shift and stay in the continuing choice lation (EV: probability × value); 2) somatic marker system Mean probability of shift and stay in the continuing may only contribute to generating subjective feelings choice. After choosing the (bad) decks A or B, 65% of par- (consciousness), and may not be immediately related to ticipants remain on these two decks, with only 35% shifting decision guidance; 3) the operation of the somatic marker to the (good) decks C and D (t (47) = 8.60, p < .01). On the may be involved in gathering the long-term memory, but other hand, when subjects choose decks C and D, the proba- may not globally direct choice behavior in situations of bilities of them shifting or staying in their next selection was a high uncertainty. roughly 50/50 probability of selecting deck A and B versus C and D (t (47) = 1.71, p = .09). The Somatic Marker Hypothesis also posits that somatic markers guide advantageous behavior in a non-conscious manner. However, questionnaire results suggest further [18,19,23,25,53] also demonstrated that deck B was cho- that subjects can have a clear knowledge of gain and loss sen more than deck C or D. frequency by the end of the game. This perception eventu- ally determines their choice patterns. A similar finding Furthermore, in Peters and Slovic study [54], the modified was obtained by Maia and McClelland [34], demonstrat- IGT study also demonstrated that deck B ($ -250) and D ing a possible "conscious" knowledge of EV in the IGT. ($ +250) possessed the inversed expected values, but with nearly equal attraction to subjects. This may imply that In the SGT, normal subjects were stuck with the influence the expected value does not guide decision makers to of gain-loss frequency without shifting to EV throughout approach the beneficial choice in these dynamic games. the entire session. If normal subjects cannot resist the On the other hand, in the modified IGT, deck C contained influence of gain-loss frequency, ventromedial prefrontal high-frequency gain (8 out of 10 trials) over deck D (5 out patients would have increased vulnerability to the effect of of 10 trials), but with nearly equal expected value ($ +300 immediate gains and losses. This prediction seems to be in for C vs. $ +250 for D). However, subjects preferred deck line with disinhibition theory that suggests that socially C rather than deck D significantly. The gain-loss frequency dysfunctional patients with prefrontal damage have diffi- seems to be more reasonable than EV in explaining these culty avoiding a punishment-associated stimulus when observations. Ahn et al. [55] confirmed the present find- that stimulus was previously associated with a reward ing by comparing the decision learning models for IGT [13,15,16,56]. If this is true, then both normal subjects and SGT respectively. and ventromedial prefrontal patients will not be respon- sive to the long-term dimension. Questionnaire data in this study also indicate a novel phe- nomenon, namely, the "money account illusion". Sub- Supposing this is the case, the facilitative effect of somatic jects ignored the EV dimension and miscounted the markers did not induce a learning effect or an advanta- money amount in terms of the strength of frequency, spe- geous shifting behavior in the long run in the SGT, as con- cifically, frequently receiving gains will leave an overall sistently proposed by IGT researchers. Careful analysis of impression of large accumulated monetary outcome for a experimental results obtained in this study identified a deck than when gains are infrequently received. subordinate phenomenon in that deck A was chosen Page 6 of 10 (page number not for citation purposes) Behavioral and Brain Functions 2008, 4:13 http://www.behavioralandbrainfunctions.com/content/4/1/13 Subject m Figure 4 emory assessment in the Soochow Gambling Task Subject memory assessment in the Soochow Gambling Task. Forty-eight subjects were required to report their behavior and preferences after completion of the game. (4A) Most of the sample (36 subjects) had vivid impressions for high- 2 2 frequency gain decks (A+B) (x (1) = 12.00, p < .01), (4B) but not high-frequency loss (C+D) (x (1) = 0.00, p = 1.00). (4C) Additionally, most subjects (31 subjects recognized decks A and B) possessed a clear (but wrong) image for the overall mone- 2 2 tary gain (x (1) = 4.08, p < .05); (4D) but a blurred image for overall monetary loss (x (1) = .08, p = .77). (4E) After completing the game, subjects erroneously equated the high-frequency gains as the overall advantage. Thirty one out of forty eight subjects indicated the favorable choice to decks A and B rather than C and D (x (1) = 4.08, p < .05). (4F) Most unfavorable choice they had memorized were decks C and D which possessed the high-frequency loss and positive expected value (x (1) = 8.33, p < .01). Page 7 of 10 (page number not for citation purposes) Behavioral and Brain Functions 2008, 4:13 http://www.behavioralandbrainfunctions.com/content/4/1/13 more often than deck B, despite both decks possessing the Competing interests same gain-loss frequency. The immediate value of gain The authors certify that the information listed above is and loss can also alter slightly decision-makers' choices complete to the best of our original research. The authors and is worthy of further exploration. declare that they have no competing interests. EV and gain-loss frequency were seriously confounded in Authors' contributions the administration of the IGT. Selection of good decks by This design of Soochow Gambling Task was constructed normal subjects cannot only be attributed to the effect of by YC and CH, both have made equal contributions to EV, but must also be explained in terms of gain-loss fre- thought, data interpretation, drafting the key concept, and quency. The Somatic Marker Hypothesis was further rein- revising it critically. JT provided several times of critical forced by adopting the principal findings from the IGT review to organize the whole picture for this work and experiment as important supporting data, i.e., somatic revised this article significantly. PL worked on the compu- markers predispose normal subjects to search for EV in the terization of task, consulting of data analysis. SY and JC long run. This study explored separately the relative con- participated in some discussion and provided some valu- tribution of EV and gain-loss frequency in the SGT, a mod- able viewpoints to improve the meaning of data and read- ified gambling task. Experimental results indicated that ability. All authors gave final approval of the version to be immediate reinforcement overrides the effect of EV. Crone published. et al. [57] also identified a local preference for gain-loss frequency; however, EV dominated gain-loss frequency in Acknowledgements The authors would like to thank the National Science Council of the their modified IGT. In contrast to the predictions based on Republic of China, Taiwan, for financially supporting this research under Somatic Marker Hypothesis, experimental data in this Contract No. NSC 94-2413-H031-002. Professors Keng-Chen Liang and study indicate that normal subjects were primarily guided Daisy Hung are appreciated for their valuable courses for introduction of by the effect of gain-loss frequency rather than EV. This Somatic Marker Hypothesis. Dr. Ian Tomb and Professor Marc Hauser are finding, although differing from the simple explanation also appreciated for kindly providing the original instruction of IGT. Thanks provided by the Somatic Marker Hypothesis, is consistent also extend to the editors and reviewers of several journals over the past with the view of behavioral [1-5] and affective [5,20,58- 4 years for their valuable comments. 62] decision literature, indicating that normal individuals are often short-sighted when making decisions in stock References 1. Hastie R, Dawes RM: Rational Decision in an uncertainty world: market or real life [41,63,64]. The psychology of judgment and decision making. London: Sage Publications; 2001. Conclusion 2. Kahneman D: Maps of bounded rationality: Psychology for behavioral economics. Amer Econ Rev 2003, 93:1449-1475. A serious confounding effect in IGT was demonstrated by 3. Kahneman D, Tversky A: Prospect theory: An analysis of deci- the "prominent deck B phenomenon". This recently-dis- sion under risk. Econometrica 1979, 47:263-291. 4. Tversky A, Kahneman D: The framing of decisions and the psy- covered phenomenon may imply that gain-loss frequency chology of choice. Science 1981, 211:453-458. rather than EV dominates the choice behavior of normal 5. Panksepp J: Damasio's Error. Consciousness and Emotion 2003, decision makers under uncertainty. In the original IGT, 4:111-134. 6. Lichtenstein S, Slovic P, Zink DJ: Effect of instruction in expected these two factors were seriously confounded, and thus this value on optimality of gambling decisions. J Exp Psychol Gen study proposed a modified task SGT to clarify the relative 1969, 79:236-240. contribution of these two factors. The experimental results 7. Damasio AR: Descartes' error: emotion, reason, and the human brain. New York: G. P. Putnam's Sons; 1994. indicated that subject selection patterns were mostly pre- 8. Bechara A, Damasio AR, Damasio H, Anderson SW: Insensitivity to dicted based on gain-loss frequency. Immediate gain may future consequences following damage to human prefrontal cortex. Cognition 1994, 50:7-15. increase the probability of continuing to stick on the cho- 9. Bechara A, Damasio H, Tranel D, Damasio AR: Deciding advanta- sen deck, which is consistent with most behavioral-analy- geously before knowing the advantageous strategy. Science sis and decision literatures. The observations of this study 1997, 275:1293-1295. 10. Leland J, Grafman J: Experimental tests of Somatic Marker indicate that EV is not predictive for choice behaviors as hypothesis. Games Econ Behav 2005, 52:386-409. suggested by IGT. It seems that immediate gain eventually 11. Tomb I, Hauser M, Deldin P, Caramazza A: Do somatic markers mediate decisions on the gambling task? Nat Neurosci 2002, resulted in a long-term loss. The basic assumption of IGT 5:1103-1104. is thus no longer plausible to assure the foresighted deci- 12. Suzuki A, Hirota A, Takasawa N, Shigemasu K: Application of the sion makers in the Iowa Gambling Task. somatic marker hypothesis to individual differences in deci- sion making. Biol Psychol 2003, 65(1):81-88. 13. Rolls ET: Emotion Explained. New York: Oxford University Press; Abbreviations EV: Expected Value; IGT: Iowa Gambling Task; SGT: Soo- 14. Dunn BD, Dalgleish T, Lawrence AD: The somatic marker hypothesis: A critical evaluation. Neurosci Biobehav Rev 2006, chow Gambling Task; SMH: Somatic Marker Hypothesis 30(2):239-271. 15. Rolls ET: The brain and emotion. New York: Oxford University Press; 1999. Page 8 of 10 (page number not for citation purposes) Behavioral and Brain Functions 2008, 4:13 http://www.behavioralandbrainfunctions.com/content/4/1/13 16. Kringelbach ML, Rolls ET: The functional neuroanatomy of the 37. Bechara A, Tranel D, Damasio H, Damasio AR: Failure to respond human orbitofrontal cortex: evidence from neuroimaging autonomically to anticipated future outcomes following and neuropsychology. Prog Neurobiol 2004, 72:341-372. damage to prefrontal cortex. Cereb Cortex 1996, 6:215-225. 17. Lin CH, Chiu YC, Huang JT: Growing evidence of prominent 38. Bechara A, Damasio H, Tranel D, Anderson SW: Dissociation Of deck B phenomenon in Iowa Gambling Task. Society of Neur- working memory from decision making within the human oeconomics 5th Annual Meeting: 2007; Boston, MA 2007. prefrontal cortex. J Neurosci 1998, 18(1):428-437. 18. Bark R, Dieckmann S, Bogerts B, Northoff G: Deficit in decision 39. Bechara A, Tranel D, Damasio H, Damasio AR: Characterization making in catatonic schizophrenia: an exploratory study. Psy- of the decision-making deficit of patients with ventromedial chiatry Res 2005, 134(2):131-141. prefrontal cortex lesions. Brain 2000, 123:2189-2202. 19. Wilder KE, Weinberger DR, Goldberg TE: Operant conditioning 40. Chiu YC, Lin CH, Huang JT, Lin S, Lee PL, Hsieh JC: Immediate gain and the orbitofrontal cortex in schizophrenic patients: unex- is long-term loss: Are there foresighted decision makers in pected evidence for intact functioning. Schizophr Res 1998, Iowa gambling task? Society for Neuroeconomics 3rd Annual Meeting: 30(2):169-174. September 15–18 2005; Kiawah Island, South Carolina, U.S.A. 2005. 20. O'Carroll RE, Papps BP: Decision making in humans: the effect 41. Chiu YC, Lin CH, Lin S, Huang JT: Reexamining the Effect of of manipulating the central noradrenergic system. J Neurol Long-term Outcome and Gain-loss Frequency: From Uncer- Neurosurg Psychiatry 2003, 74(3):376-378. tainty to Certainty. Society for Neuroeconomics 4th Annual Meeting: 21. Crone EA, Somsen RJM, van Beek B, van Der Molen MW: Heart 2006; Park City, Utah, U.S.A. 2006. rate and skin conductance analysis of antecendents and con- 42. Mitropoulos A: An experiment on the value of structural infor- sequences of decision making. Psychophysiology 2004, mation in a 2 × 2 repeated game. Econ Letters 2003, 78:27-32. 41:531-540. 43. Colman AM: Oxford Dictionary of Psychology. 1st edition. New 22. Overman WH, Frassrand K, Ansel S, Trawalter S, Bies B, Redmond York: Oxford University Press; 2003. A: Performance on the IOWA card task by adolescents and 44. Herrnstein RJ: Formal properties of the matching law. J Exp adults. Neuropsychologia 2004, 42(13):1838-1851. Anal Behav 1974, 21:159-164. 23. Ritter LM, Meador-Woodruff JH, Dalack GW: Neurocognitive 45. Rachlin H: Introduction to modern behaviorism. 3rd edition. measures of prefrontal cortical dysfunction in schizophrenia. New York: W. H. Freeman and Company; 1991. Schizophr Res 2004, 68(1):65-73. 46. Schwartz B, Wasserman EA, Robbins SJ: Psychology of learning 24. Rodriguez-Sanchez JM, Crespo-Facorro B, Perez-Iglesias R, Gonzalez- and Behavior. 5th edition. New York: W. W. Norton & Company; Blanch C, Alvarez-Jimenez M, Llorca J, Vazquez-Barquero JL: Pre- 2002. frontal cognitive functions in stabilized first-episode patients 47. Catania AC, Sagvolden T: Preference for free choice over forced with schizophrenia spectrum disorders: a dissociation choice in pigeons. J Exp Anal Behav 1980, 34:77-86. between dorsolateral and orbitofrontal functioning. Schizophr 48. Catania AC: Learning. 4th edition. NJ: Prentice Hall: Upper Saddle Res 2005, 77(2–3):279-288. River; 1998. 25. Toplak ME, Jain U, Tannock R: Executive and motivational proc- 49. Domjan M: The priciples of learning and behavior. fourth edi- esses in adolescents with Attention-Deficit-Hyperactivity tion. Belmont, CA: Wadsworth/Thompson; 2002. Disorder (ADHD). Behav Brain Funct 2005, 1(1):8. 50. Sagvolden T, Aase H, Zeiner P, Berger D: Altered reinforcement 26. Fernie G, Tunney RJ: Some decks are better than others: the mechanisms in attention-deficit/hyperactivity disorder. effect of reinforcer type and task instructions on learning in Behav Brain Res 1998, 94(1):61-71. the Iowa Gambling Task. Brain Cogn 2006, 60(1):94-102. 51. Johansen EB, Killeen PR, Sagvolden T: Behavioral variability, elim- 27. Caroselli JS, Hiscock M, Scheibel RS, Ingram F: The simulated gam- ination of responses, and delay-of-reinforcement gradients bling paradigm applied to young adults: an examination of in SHR and WKY rats. Behav Brain Funct 2007, 3:60. university students' performance. Appl Neuropsychol 2006, 52. Ferster CB, Skinner BF: Schedules of reinforcement. MA: Pren- 13(4):203-212. tice-Hall, Inc; 1957. 28. Martino DJ, Bucay D, Butman JT, Allegri RF: Neuropsychological 53. Shurman B, Horan WP, Nuechterlein KH: Schizophrenia patients frontal impairments and negative symptoms in schizophre- demonstrate a distinctive pattern of decision-making nia. Psychiatry Res 2007. impairment on the Iowa Gambling Task. Schizophr Res 2005, 29. Lin CH, Chiu YC, Lee PL, Hsieh JC: Is deck B a disadvantageous 72(2–3):215-224. deck in the Iowa Gambling Task? Behav Brain Funct 2007, 3:16. 54. Peters E, Slovic P: The springs of action: Affective and analyti- 30. Sevy S, Burdick KE, Visweswaraiah H, Abdelmessih S, Lukin M, cal Information processing in choice. Pers Soc Psychol Bull 2000, Yechiam E, Bechara A: Iowa gambling task in schizophrenia: a 26:1465-1475. review and new data in patients with schizophrenia and co- 55. Ahn WY, Busemeyer JR, Wagenmakers EJ, Stout JC: Comparison of occurring cannabis use disorders. Schizophr Res 2007, 92(1– decision learning models using the generalization criterion 3):74-84. method. Cogn Sci 2008 in press. Accepted pending revisions 31. Johnson CA, Xiao L, Palmer P, Sun P, Wang Q, Wei Y, Jia Y, Grenard 56. Rolls ET, Hornak J, Wade D, McGrath J: Emotion-related learning JL, Stacy AW, Bechara A: Affective decision-making deficits, in patients with social and emotional changes associated linked to a dysfunctional ventromedial prefrontal cortex, with frontal lobe damage. J Neurol Neurosurg Psychiatry 1994, revealed in 10th grade Chinese adolescent binge drinkers. 57:1518-1524. Neuropsychologia 2008, 46(2):714-726. 57. Crone EA, van der Molen MW: Developmental changes in real 32. Chiu YC, Lin CH: Is deck C an advantageous deck in the Iowa life decision making: performance on a gambling task previ- Gambling Task? Behav Brain Funct 2007, 3(37):. ously shown to depend on the ventromedial prefrontal cor- 33. Lin CH, Chiu YC, Lee PL, Hsieh JC: The preferences of decision- tex. Dev Neuropsychol 2004, 25(3):251-279. making in IOWA Gambling Task: The testing of frequency 58. Berridge K: Irrational Pursuits: Hyper-Incentives from a Vis- effect and long-term outcomes in Somatic Marker Hypothe- ceral Brain. In The psychology of economics decisions: Volume I: Ration- sis. 2nd Conference of NeuroEconomics: 2004; Münster, Germany 2004. ality and Well-Being Edited by: Brocas I, Carrillo J. New York: Oxford 34. Maia TV, McClelland JL: A reexamination of the evidence for the University Press; 2003:17-40. somatic marker hypothesis: what participants really know in 59. Baumeister R: The psychology of irrationality: Why people the Iowa gambling task. Proc Natl Acad Sci USA 2004, make foolish, self-defeating choices. In The psychology of econom- 101(45):16075-16080. ics decisions: Volume I: Rationality and Well-Being Edited by: Brocas I, 35. Bowman CH, Evans CEY, Turnbull OH: Artificial time constraints Carrillo J. New York: Oxford University Press; 2003. on the Iowa Gambling Task: The effects on behavioral per- 60. Heims HC, Critchley HD, Dolan R, Mathias CJ, Cipolotti L: Social formance and subjective experience. Brain Cogn 2005, and motivational functioning is not critically dependent on 57:21-25. feedback of autonomic responses: neuropsychological evi- 36. Bechara A, Damasio H, Damasio AR, Lee GP: Different contribu- dence from patients with pure autonomic failure. Neuropsych- tions of the human amygdala and ventromedial prefrontal ologia 2004, 42(14):1979-1988. cortex to decision-making. J Neurosci 1999, 19:5473-5481. 61. North NT, O'Carroll RE: Decision making in patients with spi- nal cord damage: afferent feedback and the somatic marker hypothesis. Neuropsychologia 2001, 39(5):521-524. Page 9 of 10 (page number not for citation purposes) Behavioral and Brain Functions 2008, 4:13 http://www.behavioralandbrainfunctions.com/content/4/1/13 62. Ross D, Spurrett D, Vuchinich R: The behavioral economics and neuroeconomics disordered gambling: A policy-focused sur- vey of research. A report prepared for the South African Responsible Gambling Foundation. Cape Town 2006:1-255. 63. Peterson R: Inside the Investor's Brain: The Power of Mind Over Money. SA: John Wiley & Sons; 2007. 64. Lin CH, Chiu YC, Huang JT: Is decision-maker sensitive to expected value in the dynamic-uncertain gambles? 5th Annual Meeting of Society for Neuroeconomics: 2007; Boston, MA 2007. Publish with Bio Med Central and every scientist can read your work free of charge "BioMed Central will be the most significant development for disseminating the results of biomedical researc h in our lifetime." Sir Paul Nurse, Cancer Research UK Your research papers will be: available free of charge to the entire biomedical community peer reviewed and published immediately upon acceptance cited in PubMed and archived on PubMed Central yours — you keep the copyright BioMedcentral Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp Page 10 of 10 (page number not for citation purposes)

Journal

Behavioral and Brain FunctionsSpringer Journals

Published: Mar 19, 2008

There are no references for this article.