Running head: Cognitive decision models in cocaine abusers. Cognitive modeling analysis of the decision-making processes used by cocaine abusers
|
|
- Oliver Walsh
- 5 years ago
- Views:
Transcription
1 Cognitive Decision Models 1 Running head: Cognitive decision models in cocaine abusers Cognitive modeling analysis of the decision-making processes used by cocaine abusers Julie C. Stout, Jerome R. Busemeyer and Anli Lin Indiana University Steven R. Grant, Katherine R. Bonson National Institutes of Drug Abuse September 30, 2002 Send correspondence to: Dr. Julie C. Stout Department of Psychology, Indiana University 1101 East 10 th Street Bloomington IN phone: (812) jcstout@indiana.edu word count: 3401
2 Cognitive Decision Models 2 Abstract This article examines the theoretical basis of decision-making deficits exhibited by cocaine abusers using a laboratory decision-making task first described by Bechara, Damasio, Damasio, and Anderson (1994). A total of 12 male cocaine abusers and 14 controls performed the task, and the cocaine group performed significantly worse than the controls. A cognitive modeling analysis (Busemeyer & Stout, 2002) was used to estimate three parameters that measure importance of the cognitive, motivational, and response processes for determining the observed performance deficit. The results of this analysis indicated that motivational and choice consistency factors, but not learning/memory were mainly responsible for the decision-making deficit of the cocaine abusers in this task.
3 Cognitive Decision Models 3 Cognitive modeling analysis of the decision-making processes used by cocaine abusers Drug abusers choose the immediate rewards of using drugs despite negative longer-term consequences. Laboratory based decision tasks are being actively pursued as scientific tools to uncover how cognition and motivation interact with contingencies in the environment to create and sustain drug abuse in some individuals. For example, in the simulated gambling task first described by Bechara, Damasio, Damasio, and Anderson (1994), hereafter referred to as the card deck gambling task, participants select cards from four decks which have unknown contingencies. Participants are told to try to maximize their winnings, but because the contingencies of the decks are not made explicit, they must sample from the decks to learn the associations between particular decks and the win/loss outcomes. Several studies have shown that drug abusers perform more poorly on this gambling task than do comparison subjects, choosing relatively more from disadvantageous and less from advantageous decks (Bechara & Damasio, 2002; Bechara, Dolan, & Hindes, 2002; Grant, Contoreggi, & London, 2000). Other decision tasks, such as a betting and delay discounting (Kirby, Petry, & Bickel, 1999; Petry, Bickel, & Arnett, 1998; Rogers et al., 1999) have been used to isolate the conditions under which drug abusers show differences in decision behavior relative to controls. These decision tasks have also been used to suggest neurobiological substrates of drug abuse, and such work has generally supported the view that prefrontal cortical regions, especially the orbitofrontal cortex with its associated cortical and subcortical systems, may be important
4 Cognitive Decision Models 4 in drug abuse (Grant, Contoreggi et al. 2000; London, Ernst et al. 2000; Bechara and Damasio 2002). Decision tasks are complex and have unobservable components. Thus, failure to perform well on these tasks can be due to any of a number of causes. For example, poor performance on the gambling task may be due to poor learning of the contingencies, differences in the motivational significance of wins or losses, or even from an impulsive or erratic response style which can reduce the fidelity between one s knowledge about the contingencies and the selections that he or she makes. To uncover the sources of performance variations across decision tasks and across individuals, some studies use variations in the task design, for example reversing the patterns of wins and losses (Bechara, Dolan et al., 2002) to isolate possible explanations of choice behavior. Another approach is to use additional measures, such as indices of personality or cognitive ability, to identify individual differences that are associated with decision task performance (Petry, Bickel et al., 1998; Mazas, Finn et al., 2000; Bechara and Damasio, 2002; Bechara, Dolan et al., 2002). These approaches have been informative, but remain open to alternative interpretations because they rely on face validity rather than direct theoretical evidence. Mathematical psychologists are developing new theoretical tools that provide rigorous evidence for identifying the psychological processes underlying complex behaviors in cognitive tasks (Ratcliff, Thapar et al., 2001; Zaki and Nosofsky, 2001; Busemeyer and Stout, 2002; Reifer, Knapp et al., 2002). For this approach, simple mathematical models of a target task are developed and tested, and the parameters estimated from these cognitive models are used to describe the relative influences of task
5 Cognitive Decision Models 5 features on an individual s performance. Using this approach, individual difference measures of the relevant cognitive processes are extracted from the very same performance data that one desires to explain. This approach has been used for many years, for example, in signal detection theory, to estimate discriminability and bias from performance on detection tasks. Recently, we developed a cognitive model of decision-making that reveals the underlying psychological processes influencing performance on the card deck gambling task (Busemeyer & Stout, 2002). This initial application was limited to data collected from patients Huntington Disease, which is a progressive dementing illness. The purpose of this report is to provide new empirical tests of the model by evaluating it with data collected from cocaine abusers. For this study, we first fit candidate decision models to the data from individual subjects. Next we identified the best fitting model, and then we used that model to estimate the parameters for the individual subject data. The parameters of the model provide indices of the contribution of the psychological processes that explain the decision-making deficits exhibited by the drug abusers. These derived measures can also be used to determine the decisional characteristics that are most strongly associated with other individual difference variables, such as personality measures or indices of drug abuse severity. The remainder of this article is organized as follows. First we review the results of the empirical study comparing performance of cocaine abusers and controls on the simulated gambling task. Second, we apply our cognitive modeling approach to assess individual differences in this data. Finally, we compare the results obtained in our study
6 Cognitive Decision Models 6 of Huntington disease (Stout, Rodawalt, & Siemers, 2001) with the new results found with cocaine abuse populations. Method Participants and Procedure As part of a study in the NIDA Intramural Program, 14 control (age 30.0 ± 6.1, 100% male, estimated IQ ± 7.62) and 12 cocaine-abusing (age 36.9 ± 10.3, 79% male, 93.7 ± 10.3) individuals participated. The cocaine group was older (t(24) = -3.04, p <.01) and had lower IQs (t(24) = 3.22, p <.01) than did the control group. Exclusion criteria were history of significant medical illness or head trauma with loss of consciousness, Diagnostic and Statistical Manual (DSM) Axis I diagnosis other than Substance Abuse and Axis II diagnosis other than Borderline or Antisocial Personality Disorder. The cocaine group reported current regular cocaine abuse, and all had past experience with other drugs of abuse (stimulants, depressants, marijuana, opiates. One had a history alcohol abuse. IQ was estimated using the Shipley IQ Scale (Zachary, 1986). Participants were also in a PET study of the gambling task, although the Positron Emission Tomography (PET) results are not included in this report. The study was approved by the NIDA Institutional Review Board and all participants gave written informed consent. Procedure Participants were tested while seated in a comfortable chair in front of a computer monitor. Four card decks labeled (A, B, C, D) were displayed on the monitor, and participants were told to accumulate as much (play) money as possible by picking cards
7 Cognitive Decision Models 7 from the decks using the keyboard. Also on the display monitor, participants viewed levels of their winnings shown as stacks of coins. Decks differed with respect to the payoff for each card selection and the frequency and severity of penalties. Each selection from Decks A and B yielded $100, and each selection from Decks C and D yielded $50 per selection. These winnings were paired with losses on many cards such that Decks A and B were disadvantageous overall, leading to a net loss of $250 per 10 cards, whereas Decks C and D were advantageous overall, leading to a net gain of $250 overall. Decks A and C yielded lower frequency, larger magnitude losses, whereas Decks B and D yielded smaller, more frequent losses (for additional details, see Grant et al., 2000). Cognitive Modeling of Gambling Task To begin, we applied several dozen candidate formal cognitive models to the card task data. We then chose the model that maximized model fit indices across all participants data (a detailed example of this process is available in Busemeyer and Stout, 2002). Model fits were quantified for individual subjects by the difference, termed G 2, between the amounts of information accounted for in the cognitive model in comparison to a baseline model. The baseline model perfectly reproduces each individual s marginal choice probabilities, but it assumes a stationary multinomial process. Both the baseline model and the cognitive model have the same number of parameters (three). In order to improve over the baseline model, the cognitive model must account for the dynamic and sequential properties of each individual s data. Here, we present only the expectancyvalence model, which yielded the best fit for the data i. The expectancy - valence model yields three parameters that describe motivational, learning/memory, and choice
8 Cognitive Decision Models 8 consistency components of the data. Parameter values are estimated separately for each participant by solving for the parameter values that, according to the model, maximizes the likelihood of each participant's data. Thus, for each participant, the modeling process yields three theoretically derived parameter estimates of the three hypothesized psychological processes that influence choice behavior. These parameter estimates can be used as additional individual difference measures and correlated with other measures such as age, IQ, and severity of drug use. The mathematical details are given in the Appendix. The three individual difference parameters are described next. Definition of the motivational parameter, w: According to the expectancy-valence model, the decision maker experiences an affective reaction, called a valence, to the consequences (i.e., wins, losses) produced by each selection. Formally, this valence is represented by a weighted average (thus, w) of the losses and wins produced by the chosen deck. The motivational parameter indexes relative attention of the decision maker to the loss versus the win experienced with each selection. For example, cocaine abusers may persist in choosing from the disadvantageous decks because they are insensitive to the large losses produced by these decks. Such insensitivity to loss is indicated by smaller values of w. Definition of the learning/memory parameter, a: Based on past experience, the decision maker forms expectations about the consequences delivered by each deck. Whenever a deck is chosen, the newly experienced valence produced by that choice modifies the expectancy for that deck. Formally, the new expectancy for a deck is a weighted average of the previous expectancy and the new valence. The weight allocated to the newly experienced valence is denoted a, which is an updating rate parameter.
9 Cognitive Decision Models 9 Large values of a indicate strong recency effects and more rapid forgetting of past outcomes. Small values of a indicate the persistence of influences over longer spans of selections. For example, cocaine abusers may have a large learning rate which would cause them to forget the infrequently occurring large losses produced by these decks, and thus continue to choose disadvantageously. Consistency of choice behavior,?: The decision maker's choice on each trial is based not only on the expectancies produced by each deck, but also on the consistency with which the decision maker applies those expectancies when making the selections. This consistency is denoted θ. In effect, according to the expectency-valence model, the probability of choosing a deck is a ratio of the strength of that deck relative to the sum of the strengths of all the decks. Furthermore, the strength of a given deck is determined by the expectancy for that deck multiplied by the consistency parameter. When the consistency is very low, choices are random, reckless, impulsive, and independent of the expectancies; when consistency is very high, the deck that has the maximum expectancy will almost certainly be chosen on each trial. For example, cocaine abusers may fail to select a high proportion of advantageous cards because they respond unreliably to their expectations, and their choices are mostly random guesses, an effect that would lead to lower values on the consistency parameter, θ. Results Gambling Task Performance The control group gradually learned to choose from the advantageous decks, whereas the cocaine group chose nearly equally from the advantageous and disadvantageous decks (see Figure 1, jagged curves). This group effect remained after
10 Cognitive Decision Models 10 covarying out the effects of age and IQ. The cocaine group selected significantly fewer advantageous cards (analysis of covariance, F(1,22) = 7.8, MSE = 2.7, p =.01). Cognitive modeling results For comparison, Figure 1 shows the choice predictions derived from the expectancy-valence model with the observed proportion of choices (jagged curves). As can be seen in this figure, the expectancy-valence model tracks the observed trends fairly closely. The percentages of variance predicted by the model for the averaged results are 82% and 30% for the control and cocaine abuse groups respectively. For the individual analyses, a total of 81% of the participants (13 out of 14 controls, and 8 out of 12 cocaine abusers) produced improved fit statistics, indicating improved fit of the expectancy valence model over the baseline model ii. This percentage is significantly greater than expected by chance (z = 3.14, p <.01). All of the 5 subjects with poor model fits performed poorly on the task, choosing nearly equally from the advantageous and disadvantageous decks. A comparison of the groups on the three model parameters indicated group effects on the motivational and consistency, but not learning/memory parameters (see Table 1). Although the learning/memory parameter did not differ significantly across the two groups (t(17.41) = -.977, p =.34), the variance within groups differed (F = 6.77, p =.016, based on Levene's test for equality of variances). The variance was significantly larger for the cocaine abuse group, which produced both very low and very high learning/memory parameters. The motivational parameter was significantly lower for the cocaine abuse group than the controls (t(24) = 3.40, p =.002), indicating that the card selections of the cocaine abuse group were less influenced by losses than were controls. The consistency
11 Cognitive Decision Models 11 parameter for the cocaine abusers was significantly smaller than the control group (t(24) = 2.53, p =.018). This indicates that the card selections of the cocaine abusers were more erratic (i.e., less related to their overall expectancies of deck outcome). To ensure that these differences in parameters generated from the model were not explained by the group differences in age and IQ, we also computed an analysis of covariance, with age and IQ as the covariates. Results were similar, with a significant difference between the groups in the motivational parameter, a trend for a group difference in the choice consistency parameter, but no apparent difference in the learning/memory parameter. Association of model parameters with age and IQ Separate multiple regression analyses were performed to determine the magnitude of the independent effects of age and IQ on each of the model parameters. Neither age nor IQ had significant effects on the learning parameter. For the motivational parameter, although IQ was not significant, there was a marginal effect of age (β = -.42, t = -2.00, p =.057). For the consistency parameter, although age was not significant, there was a significant effect of IQ (β =.54, t = 2.72, p =.012). Utility of model parameters for predicting group membership A discriminant function analysis, used to determine the extent to which the three parameters could predict group membership, was statistically significant (Wilks Lambda =.506, chi square = 15.2, p =.002); all of the drug abuse participants were correctly classified; 11 of the 14 controls were correctly classified, so the overall rate of correct classification was 89%. The motivational parameter produced the largest standardized coefficient (.88), and the consistency parameter was next (.71), and the learning/memory parameter contributed very little (-.05).
12 Cognitive Decision Models 12 Discussion Basic Findings Cocaine abusers failed to learn to choose from the advantageous decks in the card deck gambling task, whereas controls rapidly learned to do so. The expectancy-valence model provided a better fit than a baseline model for 81% of the participants. The present study identified which of three possible psychological explanations were responsible for this deficit. The parameter estimates obtained from the expectancy-valence model yielded the following conclusions. First, the decision making of the cocaine abusers on the task was less responsive to losses than was the decision making of the controls. This suggests differences in motivational processes are a likely source of the differences in the decision-making styles of cocaine abusers, which is consistent with several major accounts in the drug abuse literature (Kirby, Petry et al., 1999; Grant, Contoreggi et al., 2000; London, Ernst et al., 2000; Bechara and Damasio, 2002; Bechara, Dolan et al., 2002). Second, the choices of the cocaine group were more random and less consistently related to expected payoffs than the choices made by the control group. This indicates that the response process also contributes to the decision-making differences between cocaine abusers and the control group. In contrast to the motivational processes and choice consistency, the learning/memory processes did not distinguish the drug abusers from the control group. Comparison with Previous Cognitive Modeling Results Busemeyer and Stout (2002) conducted cognitive modeling analyses of Huntington participants on the card deck gambling task. Huntington Disease is a progressive degenerative neurological disease that is associated with declines in cognitive
13 Cognitive Decision Models 13 function, particularly memory, executive, and attentional functions. The decision making deficit exhibited by the Huntington disease group was explained by the learning/memory parameter a (referred to as the updating rate in Busemeyer and Stout, 2002). Differences in performance were not associated with motivational influences, w, which did not differ between the Huntington and control groups. The results from the Huntington disease study are in sharp contrast with the present study, which points strongly to motivational differences as the source of the deficit in cocaine abusers. The findings in these studies have some external validity. For example, additional evidence points to learning/memory as a likely explanation for poor performance in Huntington s disease. These individuals have a dementing disease, and memory impairment is by definition an invariant of dementia. Also, in Stout et al. (2001), performance on the gambling task was associated with other measures of learning and memory. In contrast, the finding that motivation, or the attention given to losses relative to gains, is a stronger determinant in the decision process of the cocaine abusers is consistent with the conclusions from other studies of drug abusers (Petry, Bickel et al., 1998; Kirby, Petry et al., 1999; Bechara and Damasio, 2002; Bechara, Dolan et al., 2002) iii. Interestingly, choice consistency (which was referred to as the sensitivity parameter in Busemeyer and Stout (2002)) was lower for the clinical groups in both the Huntington and cocaine abuser study. This is consistent with the possibility that reduced consistency in choice behavior may be a less specific feature in neurobehavioral disorders. The comparison of the Busemeyer and Stout (Busemeyer & Stout, 2002) and the present study illustrates an important point about what can be gained from applying a formal model to decision data from clinical groups. That is, although the behavioral
14 Cognitive Decision Models 14 results from the cocaine abusers in the current study and the Huntington participants in the previous study appear similar, these performance deficits have different underlying theoretical explanations (i.e., motivational versus learning, respectively). Such differences can be revealed and systematically studied using a cognitive modeling approach. This may be particularly valuable in describing subgroups of clinical populations that show similar deficits with different underlying causes. For example, in the recent studies by Bechara et al (2002a; 2002b), two subgroups of drug abusers are identified by their performances on a variant of the card deck gambling task, one that is hypersensitive to reward, and another group that is indifferent to winning or losing. In such cases, cognitive modeling may be useful for revealing subgroups based on parameter estimates, and for explaining these subgroup differences in terms of psychological processes. The parameter estimates can also be studied in conjunction with other individual difference variables, such as personality features or the extent of drug use. Utility of Cognitive Modeling for Studies of Drug Abuse Understanding psychological processes using cognitive modeling provides a means for building a bridge between neurophysiology and behavior (these connections are developed in more detail in Stout, Bechara, Busemeyer, and Lin, in progress). Without our cognitive modeling analysis, it would not be possible to discern whether the performance deficits in cocaine abusers in this study were due to learning/memory or motivational processes. If our results had indicated that learning/memory was responsible for the performance deficit, that finding would lead to the investigation of neurophysiological mechanisms of learning and memory. Instead, our cognitive modeling
15 Cognitive Decision Models 15 analysis identifies a motivational source for the behavioral deficit. This finding leads to investigations of the brain s reward systems, an area of intense research focus in current studies of drug abuse.
16 Cognitive Decision Models 16 References Bechara, A., & Damasio, H. (2002). Decision-making and addiction (part I): Impaired activation of somatic states in substance dependent individuals when pondering decisions with negative future consequences. Neuropsychologia, 40(10), Bechara, A., Dolan, S., & Hindes, A. (2002). Decision-making and addiction (part II): Myopia for the future or hypersensitivity to reward. Neuropsychologia, 40(10), Busemeyer, J. R., & Stout, J. C. (2002). A contribution of cognitive decision models to clinical assessment: Decomposing performance on the Bechara gambling task. Psychological Assessment, 14. Grant, S., Contoreggi, C., & London, E. D. (2000). Drug abusers show impaired performance in a laboratory test of decision making. Neuropsychologia, 38(8), Kirby, K. N., Petry, N. M., & Bickel, W. K. (1999). Heroin addicts have higher discount rates for delayed rewards than non-drug using controls. Journal of Experimental Psychology: General, 128, Petry, N. M., Bickel, W. K., & Arnett, M. (1998). Shortened time horizons and insensitivity to future consequences in heroin addicts. Addiction, 93(5), Rogers, R. D., Everitt, B. J., Baldacchino, A., Blackshaw, A. J., Swainson, R., Wynne, K., Baker, N. B., Hunter, J., Carthy, E., London, M., Deakin, J. F. W., Sahakian, B. J., & Robbins, T. W. (1999). Dissociable deficits in the decision-making cognitive of chronic amphetamine abusers, opiate abusers, patients with focal damage to prefrontal
17 Cognitive Decision Models 17 cortex, and tryptophan-depleted normal volunteers: Evidence for monoaminergic mechanisms. Neuropsychopharmacology, 20, Stout, J. C., Rodawalt, W. C., & Siemers, E. R. (2001). Risky decision making in Huntington's disease. Journal of the International Neuropsychological Society, 7, Zachary, R. A. (1986). Shipley Institute of Living Scale. Los Angeles: Western Psychological Services.
18 Cognitive Decision Models 18 Appendix This appendix summarizes the main assumptions of the expectancy - valence model of performance for the Bechara task. First we introduce some notation. The two advantageous decks are labeled D 1 and D 2, and the two disadvantageous decks are labeled, D 3 and D 4. The time index, t, is used to denote the trial number, that is the total number of cards selected by a particular decision maker at some point during training. The deck chosen on trial t is denoted D(t), so that for example D(t) = D 2 if the second advantageous deck is chosen on trial t. The reward or gain received by choosing deck D is denoted R(D), i.e., R(D) = $50 for advantageous choices, and R(D) = $100 for disadvantageous choices. The loss received from choosing deck D on trial t is denoted L(D(t)), e.g., if a disadvantageous deck is chosen on trial t=9 and a loss of $1250 occurs, then L(D(9)) = -$1250. Motivational Process. According to the expectancy-valence model, the gains and losses experienced after making a choice produce an affective reaction in the decision maker called a valence. The valence experienced after choosing deck D on trial t, denoted v(t), is represented as a weighted average of the gains or rewards, denoted R(D(T), and losses, denoted L(D(T)) : v(t) = [(1-w) R(D(t)) + w L(D(t))]. (1a) The motivation parameter, 0 < w < 1, allows for the possibility that different amounts of attention are given to the losses as compared to the gains.
19 Cognitive Decision Models 19 Learning/memory Process. The decision maker learns expectancies about the valences produced by choosing each deck during training. The expected valence for deck D i on trial t is denoted Ev[D i t]. These expectancies are updated by an adaptive learning mechanism. If deck D i is chosen on trial t, and the valence v(t) is experienced, then the expectancy for this deck is updated as follows: Ev[D i t] = (1-a) Ev[D i t-1] + a v(t). (1b) If deck D i was not chosen, then its expectancy remains unchanged. This learning model produces expectancies that are a weighted average of the past valences. The learning/memory rate parameter, 0 < a < 1, determines how much weight is given to past experience. Consistency of Choice Process. The choice made on each trial is a probabilistic function of the expectancies associated with each deck. The probability of choosing deck D i is an increasing function of the expectancy for that deck, and a decreasing function of the expectancies for the other decks. This principle is captured by the following ratio of strengths rule for choice probabilities (cf. Luce, 1959): Pr[ D i e t + 1] = 4 e j= 1 Ev[ Di t] θ ( t ) Ev[ Dj t] θ ( t ) (1c) The parameter θ(t) = (t/10) c in Equation (1c) is called the sensitivity parameter (and the coefficient c of this power function must be estimated from the data). As the sensitivity parameter increases in magnitude, the choices become more strongly dependent on the expectancies.
20 Cognitive Decision Models 20 The expectancy-valence model was compared to a baseline model, which is a statistical rather than a cognitive model. The baseline model assumes that a multinomial process generates choices with constant probabilities across trials. The probability of choosing from deck D 1 is denoted p 1, the probability of choosing from deck D 2 is denoted p 2, the probability of choosing from deck D 3 is denoted p 3, and finally the probability of choosing from deck D 4 is p 4 = 1 - (p 1 +p 2 +p 3 ). Like the cognitive model described above, the baseline model has three parameters that must be estimated from the data, p 1, p 2, and p 3. Unlike the cognitive model, this model simply assumes that the choices are independently and identically distributed across trials. Nevertheless, the baseline model is a strong competitor because it can perfectly reproduce the marginal choice probabilities, pooled across trials. Thus, a cognitive model can only perform better than this model if it succeeds in explaining how choices change over training as a function of trial-by-trial feedback. The parameters that maximize the log likelihood for each model and person were used to make model comparisons (see Busemeyer and Stout, 2002, for details about the maximum likelihood estimation procedure). The cognitive model was compared with the baseline model by computing differences in log likelihood as follows: G 2 = 2(L Cognitive - L Baseline ) (2) G 2 is a measure of improvement of the expectancy-valence model over the baseline model. If the two models being compared were nested, so that one was a special case of the other, then G 2 would be chi-square distributed with degrees of freedom equal to the difference in the number of parameters. However in this case, the models are not nested and they contain the same number of parameters, so that standard chi square tests are not
21 Cognitive Decision Models 21 applicable. Furthermore, standard non-nested model comparison measures, such as the Akaike information criterion (AIC) or the Schwartz Bayesian information criterion (BIC) produce exactly the same result as Equation (2), because there are no differences in number of parameters. Nevertheless we can continue to use the G 2 statistic as a descriptive index of model performance. Positive values of the G 2 statistic indicate that a given cognitive model performs better than the baseline model. For the control group, the mean improvement in G 2 was equal to (med = 51.13, std = 55.71); and for the cocaine abuse group, the mean was 5.27 (med = 6.36, std = 14.57). A total of 81% of the participants (13 out of 14 controls, and 8 out of 12 cocaine abusers) produced positive G 2 improvement statistics, which is significantly different from chance according to a z-test (z = 3.14, p <.01). A larger improvement was observed for the control group because this group has a larger training effect (see endnote ii). We compared dozens of other cognitive models with the expectancy valence model. In particular, we compared the heuristic strategy-switching model and the Bayesian Expected Utility model (described in Busemeyer & Stout, 2002) with the expectancy valence model. The mean G 2 for the Heuristic Strategy Learning model was and for the control and drug abuse groups, respectively. The mean G 2 for the Bayesian Expected Utility model was and for the control and drug abuse groups, respectively. Note that if the G 2 for cognitive model A exceeds the G 2 for cognitive model B, then cognitive model A performs better than cognitive model B for a particular individual. Clearly, the expectancy valence model produces a better average fit than either of these two alternative cognitive models.
22 Cognitive Decision Models 22 Author Notes Julie C. Stout, Jerome R. Busemeyer, and Anli Lin, Department of Psychology, Indiana University; Steven R. Grant and Katherine R. Bonson, Brain Imaging Center, Intramural Program at NIDA. Dr. Grant is now with the NIDA Program Office; Dr. Bonson is now at the Controlled Substances Program of the U.S. Food and Drug Administration. This research was supported by NIDA grant DA-R and by the Intramural Research Program of the National Institute of Drug Abuse. Requests for reprints should be directed to Julie Stout at
23 Cognitive Decision Models 23 Footnotes i In particular, we rigorously compared the three models discussed previously in Busemeyer and Stout (2002), called the heuristic strategy switching model, the Bayesian Expected Utility model, and the Expectancy Valence model. The latter was the best fitting model for the present data, and it was also the best fitting model for the data analyzed in Busemeyer and Stout (2002). ii The reader may question the lower percentage predicted for the drug abuse group. This necessarily follows from the fact that there is no learning curve for this group. All there is to explain are sequential dependencies over trials, so there is much less signal in the noise. We have conducted computer simulations using the same parameters for each group as we found in this paper. In the simulations, of course, all of the data is generated by the model. But the same difference in percentage predicted appears in the simulated data because of the reduced signal to noise ratio for the drug abuse group. iii More recently, we (Stout, Bechara, Busemeyer, & Lin, in progress) conducted cognitive modeling analyses of Bechara's data obtained from orbital frontal cortex damaged patients, and the results match those obtained from the cocaine abusers.
24 Cognitive Decision Models 24 Table 1: Summary of Parameter Estimates from the Expectancy Valence Model Attention Weight Update Rate Sensitivity Mean 1 Med Std Mean 2 Med Std Mean 3 Med Std Control Drug t(24) = 4.3, p = t(24) = -1.02, p = t(24) = 2.53, p =.018.
25 Cognitive Decision Models 25 Figures Figure 1. The left and right panels of Figure 1 display the results for the control and drug abuse patients, respectively. Each panel shows the proportion of participants choosing from the advantageous decks as a function of the number of cards selected. The jagged curve represents the actual data and the smooth curve represents the predictions of the model. Both curves are have been linearly filtered by a moving average window of size 7.
26 Cognitive Decision Models 26 Figure 1 Figure 1
27 Cognitive Decision Models 27 i In particular, we rigorously compared the three models discussed previously in Busemeyer and Stout (2002), called the heuristic strategy switching model, the Bayesian Expected Utility model, and the Expectancy Valence model. The latter was the best fitting model for the present data, and it was also the best fitting model for the data analyzed in Busemeyer and Stout (2002). ii The reader may question the lower percentage predicted for the drug abuse group. This necessarily follows from the fact that there is no learning curve for this group. All there is to explain are sequential dependencies over trials, so there is much less signal in the noice. We have conducted computer simulations using the same parameters for each group as we found in this paper. In the simulations, of course, all of the data is generated by the model. But the same percentage predicted appears in the simulated data because of the reduced signal to noise ratio for the drug abuse group. iii More recently, we (Stout, Bechara, Busemeyer, & Lin, in progress) conducted cognitive modeling analyses of Bechara's data obtained from orbital frontal cortex damaged patients, and the results match those obtained from the cocaine abusers.
A Contribution of Cognitive Decision Models to Clinical Assessment: Decomposing Performance on the Bechara Gambling Task
Psychological Assessment Copyright 2002 by the American Psychological Association, Inc. 2002, Vol. 14, No. 3, 253 262 1040-3590/02/$5.00 DOI: 10.1037//1040-3590.14.3.253 A Contribution of Cognitive Decision
More informationUsing Computational Models to Help Explain. Decision Making Processes of Substance Abusers
Models of Substance Abuse 1 Using Computational Models to Help Explain Decision Making Processes of Substance Abusers by Jerome R. Busemeyer Julie C. Stout & Peter Finn Indiana University January 10, 2003
More informationIndividual differences in the response to forgone payoffs: An examination of high. functioning drug abusers
Response to forgone payoffs 0 Individual differences in the response to forgone payoffs: An examination of high functioning drug abusers Eldad Yechiam, Julie C. Stout*, Jerome R. Busemeyer, Stephanie L.
More informationCharacterization of the decision-making deficit of patients with ventromedial prefrontal cortex lesions
Brain (2000), 123, 2189 2202 Characterization of the decision-making deficit of patients with ventromedial prefrontal cortex lesions Antoine Bechara, Daniel Tranel and Hanna Damasio Department of Neurology,
More informationA Comparison of Reinforcement Learning Models for the Iowa Gambling Task Using Parameter Space Partitioning. Abstract
A Comparison of Reinforcement Learning Models for the Iowa Gambling Task Using Parameter Space Partitioning Helen Steingroever, 1 Ruud Wetzels, 2 and Eric-Jan Wagenmakers 1 Abstract The Iowa gambling task
More informationRisky Decision Making Assessed With the Gambling Task in Adults with HIV
Digital Commons@ Loyola Marymount University and Loyola Law School Psychology Faculty Works Psychology 5-1-2006 Risky Decision Making Assessed With the Gambling Task in Adults with HIV David J. Hardy Loyola
More informationSomatic Marker Hypothesis. Tasha Poppa Affective Computing Guest Lecture Fall 2017
Somatic Marker Hypothesis Tasha Poppa Affective Computing Guest Lecture Fall 2017 Outline Historical context Some Useful Functional Neuroanatomy VMPFC patients: Clinical observations Somatic Marker Hypothesis
More informationBayes Factors for Reinforcement-Learning Models of the Iowa Gambling Task
STEINGROEVER, WETZELS, AND WAGENMAKERS 1 Bayes Factors for Reinforcement-Learning Models of the Iowa Gambling Task Helen Steingroever a, Ruud Wetzels b, and Eric-Jan Wagenmakers a a Department of Psychology,
More informationPSYCHOLOGICAL SCIENCE. Research Report. CONFLICT AND THE STOCHASTIC-DOMINANCE PRINCIPLE OF DECISION MAKING Adele Diederich 1 and Jerome R.
Research Report CONFLICT AND THE STOCHASTIC-DOMINANCE PRINCIPLE OF DECISION MAKING Adele Diederich 1 and Jerome R. Busemeyer 2 1 Univesitat Oldenburg, Oldenburg, Germany, and 2 Indiana University Abstract
More informationDecision-making and addiction (part II): myopia for the future or hypersensitivity to reward?
Neuropsychologia 40 (2002) 1690 1705 Decision-making and addiction (part II): myopia for the future or hypersensitivity to reward? Antoine Bechara a,, Sara Dolan b, Andrea Hindes a a Department of Neurology,
More informationEvaluating internal consistency in cognitive models for repeated choice tasks
Evaluating internal consistency in cognitive models for repeated choice tasks Manuscript submitted on February 13, 2007 Abstract: The present paper proposes and compares two new methods for evaluating
More informationConnect with amygdala (emotional center) Compares expected with actual Compare expected reward/punishment with actual reward/punishment Intuitive
Orbitofronal Notes Frontal lobe prefrontal cortex 1. Dorsolateral Last to myelinate Sleep deprivation 2. Orbitofrontal Like dorsolateral, involved in: Executive functions Working memory Cognitive flexibility
More informationSocial and Emotional Decision-making Following Frontal Lobe Injury
Neurocase, 10(5): 398 403, 2004 Copyright Taylor & Francis Inc. 1355-4795/04/1005 398$16.00 Neurocase Social and Emotional Decision-making Following Frontal Lobe Injury Luke Clark 1 and Facundo Manes 2
More informationSUPPLEMENTAL MATERIAL
1 SUPPLEMENTAL MATERIAL Response time and signal detection time distributions SM Fig. 1. Correct response time (thick solid green curve) and error response time densities (dashed red curve), averaged across
More informationAge-related quantitative and qualitative changes in decision making ability
Behavioural Neurology 19 (2008) 59 63 59 IOS Press Age-related quantitative and qualitative changes in decision making ability Valeria Isella a,, Cristina Mapelli a, Nadia Morielli a, Oriana Pelati b,
More informationSelection and Combination of Markers for Prediction
Selection and Combination of Markers for Prediction NACC Data and Methods Meeting September, 2010 Baojiang Chen, PhD Sarah Monsell, MS Xiao-Hua Andrew Zhou, PhD Overview 1. Research motivation 2. Describe
More informationIndividual differences in decision-making
Personality and Individual Differences 39 (2005) 991 998 www.elsevier.com/locate/paid Individual differences in decision-making Ingmar H.A. Franken *, Peter Muris Department of Psychology, Erasmus University
More informationAcademic year Lecture 16 Emotions LECTURE 16 EMOTIONS
Course Behavioral Economics Academic year 2013-2014 Lecture 16 Emotions Alessandro Innocenti LECTURE 16 EMOTIONS Aim: To explore the role of emotions in economic decisions. Outline: How emotions affect
More informationMaking Assessments While Taking Sequential Risks
Making Assessments While Taking Sequential Risks Avishai Wershbale (wershbal@msu.edu) & Timothy J. Pleskac (pleskact@msu.edu) Department of Psychology, Michigan State University, East Lansing, MI 48824
More informationUvA-DARE (Digital Academic Repository) Safe models for risky decisions Steingröver, H.M. Link to publication
UvA-DARE (Digital Academic Repository) Safe models for risky decisions Steingröver, H.M. Link to publication Citation for published version (APA): Steingröver, H. M. (2017). Safe models for risky decisions
More informationEvaluating generalizability and parameter consistency in learning models
1 Evaluating generalizability and parameter consistency in learning models By Eldad Yechiam Technion Israel Institute of Technology Jerome R. Busemeyer Indiana University, Bloomington IN To Appear in Games
More informationThe Somatic Marker Hypothesis: Human Emotions in Decision-Making
The Somatic Marker Hypothesis: Human Emotions in Decision-Making Presented by Lin Xiao Brain and Creativity Institute University of Southern California Most of us are taught from early on that : -logical,
More informationWORKING MEMORY AND INDIVIDUAL DIFFERENCES IN DECISION MAKING TINA L. JAMESON
WORKING MEMORY AND INDIVIDUAL DIFFERENCES IN DECISION MAKING By TINA L. JAMESON A dissertation submitted in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY WASHINGTON STATE
More informationRisk attitude in decision making: A clash of three approaches
Risk attitude in decision making: A clash of three approaches Eldad Yechiam (yeldad@tx.technion.ac.il) Faculty of Industrial Engineering and Management, Technion Israel Institute of Technology Haifa, 32000
More informationComponents of working memory and somatic markers in decision making
Psychonomic Bulletin & Review 2004, 11 (3), 515 520 Components of working memory and somatic markers in decision making TINA L. JAMESON, JOHN M. HINSON, and PAUL WHITNEY Washington State University, Pullman,
More informationUvA-DARE (Digital Academic Repository) Safe models for risky decisions Steingröver, H.M. Link to publication
UvA-DARE (Digital Academic Repository) Safe models for risky decisions Steingröver, H.M. Link to publication Citation for published version (APA): Steingröver, H. M. (2017). Safe models for risky decisions
More informationUvA-DARE (Digital Academic Repository) Safe models for risky decisions Steingröver, H.M. Link to publication
UvA-DARE (Digital Academic Repository) Safe models for risky decisions Steingröver, H.M. Link to publication Citation for published version (APA): Steingröver, H. M. (2017). Safe models for risky decisions
More informationPerceived time pressure and the Iowa Gambling Task
Judgment and Decision Making, Vol. 3, No. 8, December 2008, pp. 636 640 Perceived time pressure and the Iowa Gambling Task Michael A. DeDonno and Heath A. Demaree Department of Psychology Case Western
More informationDecision making under time pressure: An independent test of sequential sampling models
Memory & Cognition 1999, 27 (4), 713-725 Decision making under time pressure: An independent test of sequential sampling models ITIEL E. DROR Southampton University, Highfield Southampton, England JEROME
More informationFeedback Produces Divergence From Prospect Theory in Descriptive Choice Ryan K. Jessup, Anthony J. Bishara, and Jerome R.
PSYCHOLOGICAL SCIENCE Research Article Feedback Produces Divergence From Prospect Theory in Descriptive Choice Ryan K. Jessup, Anthony J. Bishara, and Jerome R. Busemeyer Department of Psychological and
More informationStudy Guide for the Final Exam
Study Guide for the Final Exam When studying, remember that the computational portion of the exam will only involve new material (covered after the second midterm), that material from Exam 1 will make
More informationSawtooth Software. MaxDiff Analysis: Simple Counting, Individual-Level Logit, and HB RESEARCH PAPER SERIES. Bryan Orme, Sawtooth Software, Inc.
Sawtooth Software RESEARCH PAPER SERIES MaxDiff Analysis: Simple Counting, Individual-Level Logit, and HB Bryan Orme, Sawtooth Software, Inc. Copyright 009, Sawtooth Software, Inc. 530 W. Fir St. Sequim,
More informationThe Role of Orbitofrontal Cortex in Decision Making
The Role of Orbitofrontal Cortex in Decision Making A Component Process Account LESLEY K. FELLOWS Montreal Neurological Institute, McGill University, Montréal, Québec, Canada ABSTRACT: Clinical accounts
More informationRegression Discontinuity Analysis
Regression Discontinuity Analysis A researcher wants to determine whether tutoring underachieving middle school students improves their math grades. Another wonders whether providing financial aid to low-income
More information3 CONCEPTUAL FOUNDATIONS OF STATISTICS
3 CONCEPTUAL FOUNDATIONS OF STATISTICS In this chapter, we examine the conceptual foundations of statistics. The goal is to give you an appreciation and conceptual understanding of some basic statistical
More informationMEA DISCUSSION PAPERS
Inference Problems under a Special Form of Heteroskedasticity Helmut Farbmacher, Heinrich Kögel 03-2015 MEA DISCUSSION PAPERS mea Amalienstr. 33_D-80799 Munich_Phone+49 89 38602-355_Fax +49 89 38602-390_www.mea.mpisoc.mpg.de
More informationFrom Habit to Addiction: A Study in Online Gambling Behavior
From Habit to Addiction: A Study in Online Gambling Behavior European Association for the Study of Gambling 8 th European Conference 14-17 September 2010 Bill Jolley, Ph.D. Norwich University Deborah Black,
More informationSUPPLEMENTARY INFORMATION
Supplementary Statistics and Results This file contains supplementary statistical information and a discussion of the interpretation of the belief effect on the basis of additional data. We also present
More information11/18/2013. Correlational Research. Correlational Designs. Why Use a Correlational Design? CORRELATIONAL RESEARCH STUDIES
Correlational Research Correlational Designs Correlational research is used to describe the relationship between two or more naturally occurring variables. Is age related to political conservativism? Are
More informationChanging expectations about speed alters perceived motion direction
Current Biology, in press Supplemental Information: Changing expectations about speed alters perceived motion direction Grigorios Sotiropoulos, Aaron R. Seitz, and Peggy Seriès Supplemental Data Detailed
More informationDissociable elements of human foresight: a role for the ventromedial frontal lobes in framing the future, but not in discounting future rewards
Neuropsychologia 43 (2005) 1214 1221 Dissociable elements of human foresight: a role for the ventromedial frontal lobes in framing the future, but not in discounting future rewards Lesley K. Fellows, Martha
More informationA Bayesian Analysis of Human Decision-Making on Bandit Problems
Steyvers, M., Lee, M.D., & Wagenmakers, E.J. (in press). A Bayesian analysis of human decision-making on bandit problems. Journal of Mathematical Psychology. A Bayesian Analysis of Human Decision-Making
More informationMODELS FOR THE ADJUSTMENT OF RATING SCALES 1
MODELS FOR THE ADJUSTMENT OF RATING SCALES 1 Gert Haubensak and Peter Petzold Justus Liebig University of Giessen, Germany gert.haubensak@psychol.uni-giessen.de Abstract In a category rating experiment
More informationA Brief Introduction to Bayesian Statistics
A Brief Introduction to Statistics David Kaplan Department of Educational Psychology Methods for Social Policy Research and, Washington, DC 2017 1 / 37 The Reverend Thomas Bayes, 1701 1761 2 / 37 Pierre-Simon
More informationRussian Journal of Agricultural and Socio-Economic Sciences, 3(15)
ON THE COMPARISON OF BAYESIAN INFORMATION CRITERION AND DRAPER S INFORMATION CRITERION IN SELECTION OF AN ASYMMETRIC PRICE RELATIONSHIP: BOOTSTRAP SIMULATION RESULTS Henry de-graft Acquah, Senior Lecturer
More informationDouble dissociation of value computations in orbitofrontal and anterior cingulate neurons
Supplementary Information for: Double dissociation of value computations in orbitofrontal and anterior cingulate neurons Steven W. Kennerley, Timothy E. J. Behrens & Jonathan D. Wallis Content list: Supplementary
More informationPutative Mechanisms Underlying Risky Decision-Making in High Risk Drivers
Putative Mechanisms Underlying Risky Decision-Making in High Risk Drivers Samantha Ashley Wells 1,2, Thomas Gordon Brown 1,2 Psychiatry Department, McGill University Psychiatry Department, Douglas Hospital
More informationIdentifying individual differences: An algorithm with application to Phineas Gage
Games and Economic Behavior 52 (2005) 373 385 www.elsevier.com/locate/geb Identifying individual differences: An algorithm with application to Phineas Gage Daniel Houser a,, Antoine Bechara b, Michael
More informationNEUROPSYCHOPHARMACOLOGY 1999 VOL. 20, NO American College of Neuropsychopharmacology
Dissociable Deficits in the Decision-Making Cognition of Chronic Amphetamine Abusers, Opiate Abusers, Patients with Focal Damage to Prefrontal Cortex, and Tryptophan-Depleted Normal Volunteers: Evidence
More informationResearch and Evaluation Methodology Program, School of Human Development and Organizational Studies in Education, University of Florida
Vol. 2 (1), pp. 22-39, Jan, 2015 http://www.ijate.net e-issn: 2148-7456 IJATE A Comparison of Logistic Regression Models for Dif Detection in Polytomous Items: The Effect of Small Sample Sizes and Non-Normality
More informationAMERICAN JOURNAL OF PSYCHOLOGICAL RESEARCH
AMERICAN JOURNAL OF PSYCHOLOGICAL RESEARCH Volume 7, Number 1 Submitted: March 11, 2011 Revisions: April 2, 2011 Accepted: April 4, 2011 Published: April 6, 2011 The Effects of Explicit and Implicit Cognitive
More informationEXPLORATION FLOW 4/18/10
EXPLORATION Peter Bossaerts CNS 102b FLOW Canonical exploration problem: bandits Bayesian optimal exploration: The Gittins index Undirected exploration: e-greedy and softmax (logit) The economists and
More informationSupplementary Information
Supplementary Information The neural correlates of subjective value during intertemporal choice Joseph W. Kable and Paul W. Glimcher a 10 0 b 10 0 10 1 10 1 Discount rate k 10 2 Discount rate k 10 2 10
More informationWeek 8 Hour 1: More on polynomial fits. The AIC. Hour 2: Dummy Variables what are they? An NHL Example. Hour 3: Interactions. The stepwise method.
Week 8 Hour 1: More on polynomial fits. The AIC Hour 2: Dummy Variables what are they? An NHL Example Hour 3: Interactions. The stepwise method. Stat 302 Notes. Week 8, Hour 1, Page 1 / 34 Human growth
More informationLearning to classify integral-dimension stimuli
Psychonomic Bulletin & Review 1996, 3 (2), 222 226 Learning to classify integral-dimension stimuli ROBERT M. NOSOFSKY Indiana University, Bloomington, Indiana and THOMAS J. PALMERI Vanderbilt University,
More informationA Cue Imputation Bayesian Model of Information Aggregation
A Cue Imputation Bayesian Model of Information Aggregation Jennifer S. Trueblood, George Kachergis, and John K. Kruschke {jstruebl, gkacherg, kruschke}@indiana.edu Cognitive Science Program, 819 Eigenmann,
More informationBayesians methods in system identification: equivalences, differences, and misunderstandings
Bayesians methods in system identification: equivalences, differences, and misunderstandings Johan Schoukens and Carl Edward Rasmussen ERNSI 217 Workshop on System Identification Lyon, September 24-27,
More informationBrain Imaging studies in substance abuse. Jody Tanabe, MD University of Colorado Denver
Brain Imaging studies in substance abuse Jody Tanabe, MD University of Colorado Denver NRSC January 28, 2010 Costs: Health, Crime, Productivity Costs in billions of dollars (2002) $400 $350 $400B legal
More informationDiscounting of Monetary and Directly Consumable Rewards Sara J. Estle, Leonard Green, Joel Myerson, and Daniel D. Holt
PSYCHOLOGICAL SCIENCE Research Article Discounting of Monetary and Directly Consumable Rewards Sara J. Estle, Leonard Green, Joel Myerson, and Daniel D. Holt Washington University ABSTRACT We compared
More informationReinforcement Learning : Theory and Practice - Programming Assignment 1
Reinforcement Learning : Theory and Practice - Programming Assignment 1 August 2016 Background It is well known in Game Theory that the game of Rock, Paper, Scissors has one and only one Nash Equilibrium.
More informationWhy so gloomy? A Bayesian explanation of human pessimism bias in the multi-armed bandit task
Why so gloomy? A Bayesian explanation of human pessimism bias in the multi-armed bandit tas Anonymous Author(s) Affiliation Address email Abstract 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
More informationAppendix: Instructions for Treatment Index B (Human Opponents, With Recommendations)
Appendix: Instructions for Treatment Index B (Human Opponents, With Recommendations) This is an experiment in the economics of strategic decision making. Various agencies have provided funds for this research.
More informationCenter for Advanced Studies in Measurement and Assessment. CASMA Research Report
Center for Advanced Studies in Measurement and Assessment CASMA Research Report Number 39 Evaluation of Comparability of Scores and Passing Decisions for Different Item Pools of Computerized Adaptive Examinations
More informationInferential Statistics
Inferential Statistics and t - tests ScWk 242 Session 9 Slides Inferential Statistics Ø Inferential statistics are used to test hypotheses about the relationship between the independent and the dependent
More informationIt is Whether You Win or Lose: The Importance of the Overall Probabilities of Winning or Losing in Risky Choice
The Journal of Risk and Uncertainty, 30:1; 5 19, 2005 c 2005 Springer Science + Business Media, Inc. Manufactured in The Netherlands. It is Whether You Win or Lose: The Importance of the Overall Probabilities
More informationA Race Model of Perceptual Forced Choice Reaction Time
A Race Model of Perceptual Forced Choice Reaction Time David E. Huber (dhuber@psych.colorado.edu) Department of Psychology, 1147 Biology/Psychology Building College Park, MD 2742 USA Denis Cousineau (Denis.Cousineau@UMontreal.CA)
More informationHeavy smokers choose large, immediate rewards with large penalties on a simulated task of gambling
Louisiana State University LSU Digital Commons LSU Master's Theses Graduate School 2003 Heavy smokers choose large, immediate rewards with large penalties on a simulated task of gambling Michael Shawn
More informationRisky Choice Decisions from a Tri-Reference Point Perspective
Academic Leadership Journal in Student Research Volume 4 Spring 2016 Article 4 2016 Risky Choice Decisions from a Tri-Reference Point Perspective Kevin L. Kenney Fort Hays State University Follow this
More informationA Decision-Theoretic Approach to Evaluating Posterior Probabilities of Mental Models
A Decision-Theoretic Approach to Evaluating Posterior Probabilities of Mental Models Jonathan Y. Ito and David V. Pynadath and Stacy C. Marsella Information Sciences Institute, University of Southern California
More informationA Race Model of Perceptual Forced Choice Reaction Time
A Race Model of Perceptual Forced Choice Reaction Time David E. Huber (dhuber@psyc.umd.edu) Department of Psychology, 1147 Biology/Psychology Building College Park, MD 2742 USA Denis Cousineau (Denis.Cousineau@UMontreal.CA)
More informationThe Effect of Frontal Lobe Stress on Gambling Task Performance: Implications for. Understanding Addictive Behavior. Jared A. Rowland, M.S.
The Effect of Frontal Lobe Stress on Gambling Task Performance: Implications for Understanding Addictive Behavior Jared A. Rowland, M.S. Dissertation submitted to the faculty of the Virginia Polytechnic
More informationThe Balloon Analogue Risk Task (BART) Differentiates Smokers and Nonsmokers
Experimental and Clinical Psychopharmacology Copyright 2003 by the American Psychological Association, Inc. 2003, Vol. 11, No. 1, 26 33 1064-1297/03/$12.00 DOI: 10.1037/1064-1297.11.1.26 The Balloon Analogue
More informationLec 02: Estimation & Hypothesis Testing in Animal Ecology
Lec 02: Estimation & Hypothesis Testing in Animal Ecology Parameter Estimation from Samples Samples We typically observe systems incompletely, i.e., we sample according to a designed protocol. We then
More informationAuthor's personal copy
Journal of Mathematical Psychology 53 (2009) 168 179 Contents lists available at ScienceDirect Journal of Mathematical Psychology journal homepage: www.elsevier.com/locate/jmp A Bayesian analysis of human
More informationUnit 1 Exploring and Understanding Data
Unit 1 Exploring and Understanding Data Area Principle Bar Chart Boxplot Conditional Distribution Dotplot Empirical Rule Five Number Summary Frequency Distribution Frequency Polygon Histogram Interquartile
More informationRisk Attitude in Decision Making: In Search of Trait-Like Constructs
Topics in Cognitive Science 3 (2011) 166 186 Copyright Ó 2011 Cognitive Science Society, Inc. All rights reserved. ISSN: 1756-8757 print / 1756-8765 online DOI: 10.1111/j.1756-8765.2010.01126.x Risk Attitude
More informationNeural Basis of Decision Making. Mary ET Boyle, Ph.D. Department of Cognitive Science UCSD
Neural Basis of Decision Making Mary ET Boyle, Ph.D. Department of Cognitive Science UCSD Phineas Gage: Sept. 13, 1848 Working on the rail road Rod impaled his head. 3.5 x 1.25 13 pounds What happened
More informationA Drift Diffusion Model of Proactive and Reactive Control in a Context-Dependent Two-Alternative Forced Choice Task
A Drift Diffusion Model of Proactive and Reactive Control in a Context-Dependent Two-Alternative Forced Choice Task Olga Lositsky lositsky@princeton.edu Robert C. Wilson Department of Psychology University
More informationCognitive modeling versus game theory: Why cognition matters
Cognitive modeling versus game theory: Why cognition matters Matthew F. Rutledge-Taylor (mrtaylo2@connect.carleton.ca) Institute of Cognitive Science, Carleton University, 1125 Colonel By Drive Ottawa,
More informationModel-Based fmri Analysis. Will Alexander Dept. of Experimental Psychology Ghent University
Model-Based fmri Analysis Will Alexander Dept. of Experimental Psychology Ghent University Motivation Models (general) Why you ought to care Model-based fmri Models (specific) From model to analysis Extended
More informationAnorexia, bulimia, and obesity: Shared decision making deficits on the Iowa Gambling Task (IGT)
Journal of the International Neuropsychological Society (2010), 16, 711 715. Copyright INS. Published by Cambridge University Press, 2010. doi:10.1017/s1355617710000354 BRIEF COMMUNICATION Anorexia, bulimia,
More informationEmpowered by Psychometrics The Fundamentals of Psychometrics. Jim Wollack University of Wisconsin Madison
Empowered by Psychometrics The Fundamentals of Psychometrics Jim Wollack University of Wisconsin Madison Psycho-what? Psychometrics is the field of study concerned with the measurement of mental and psychological
More informationThere are, in total, four free parameters. The learning rate a controls how sharply the model
Supplemental esults The full model equations are: Initialization: V i (0) = 1 (for all actions i) c i (0) = 0 (for all actions i) earning: V i (t) = V i (t - 1) + a * (r(t) - V i (t 1)) ((for chosen action
More informationAttentional Theory Is a Viable Explanation of the Inverse Base Rate Effect: A Reply to Winman, Wennerholm, and Juslin (2003)
Journal of Experimental Psychology: Learning, Memory, and Cognition 2003, Vol. 29, No. 6, 1396 1400 Copyright 2003 by the American Psychological Association, Inc. 0278-7393/03/$12.00 DOI: 10.1037/0278-7393.29.6.1396
More informationScale Invariance and Primacy and Recency Effects in an Absolute Identification Task
Neath, I., & Brown, G. D. A. (2005). Scale Invariance and Primacy and Recency Effects in an Absolute Identification Task. Memory Lab Technical Report 2005-01, Purdue University. Scale Invariance and Primacy
More informationModel evaluation using grouped or individual data
Psychonomic Bulletin & Review 8, (4), 692-712 doi:.378/pbr..692 Model evaluation using grouped or individual data ANDREW L. COHEN University of Massachusetts, Amherst, Massachusetts AND ADAM N. SANBORN
More informationThe Game Prisoners Really Play: Preference Elicitation and the Impact of Communication
The Game Prisoners Really Play: Preference Elicitation and the Impact of Communication Michael Kosfeld University of Zurich Ernst Fehr University of Zurich October 10, 2003 Unfinished version: Please do
More informationThe differential relationship between cocaine use and marijuana use on decision-making performance over repeat testing with the Iowa Gambling Task
Drug and Alcohol Dependence 90 (2007) 2 11 The differential relationship between cocaine use and marijuana use on decision-making performance over repeat testing with the Iowa Gambling Task Antonio Verdejo-Garcia
More informationExploring Complexity in Decisions from Experience: Same Minds, Same Strategy
Exploring Complexity in Decisions from Experience: Same Minds, Same Strategy Emmanouil Konstantinidis (em.konstantinidis@gmail.com), Nathaniel J. S. Ashby (nathaniel.js.ashby@gmail.com), & Cleotilde Gonzalez
More informationChapter 11. Experimental Design: One-Way Independent Samples Design
11-1 Chapter 11. Experimental Design: One-Way Independent Samples Design Advantages and Limitations Comparing Two Groups Comparing t Test to ANOVA Independent Samples t Test Independent Samples ANOVA Comparing
More informationInfluences of IRT Item Attributes on Angoff Rater Judgments
Influences of IRT Item Attributes on Angoff Rater Judgments Christian Jones, M.A. CPS Human Resource Services Greg Hurt!, Ph.D. CSUS, Sacramento Angoff Method Assemble a panel of subject matter experts
More informationCognitive Models of Choice: Comparing Decision Field Theory to the Proportional Difference Model
Cognitive Science 33 (2009) 911 939 Copyright Ó 2009 Cognitive Science Society, Inc. All rights reserved. ISSN: 0364-0213 print / 1551-6709 online DOI: 10.1111/j.1551-6709.2009.01034.x Cognitive Models
More informationRemarks on Bayesian Control Charts
Remarks on Bayesian Control Charts Amir Ahmadi-Javid * and Mohsen Ebadi Department of Industrial Engineering, Amirkabir University of Technology, Tehran, Iran * Corresponding author; email address: ahmadi_javid@aut.ac.ir
More informationLikelihood Ratio Based Computerized Classification Testing. Nathan A. Thompson. Assessment Systems Corporation & University of Cincinnati.
Likelihood Ratio Based Computerized Classification Testing Nathan A. Thompson Assessment Systems Corporation & University of Cincinnati Shungwon Ro Kenexa Abstract An efficient method for making decisions
More informationMental operations on number symbols by-children*
Memory & Cognition 1974, Vol. 2,No. 3, 591-595 Mental operations on number symbols by-children* SUSAN HOFFMAN University offlorida, Gainesville, Florida 32601 TOM TRABASSO Princeton University, Princeton,
More informationNature Neuroscience: doi: /nn Supplementary Figure 1. Task timeline for Solo and Info trials.
Supplementary Figure 1 Task timeline for Solo and Info trials. Each trial started with a New Round screen. Participants made a series of choices between two gambles, one of which was objectively riskier
More informationCHAPTER VI RESEARCH METHODOLOGY
CHAPTER VI RESEARCH METHODOLOGY 6.1 Research Design Research is an organized, systematic, data based, critical, objective, scientific inquiry or investigation into a specific problem, undertaken with the
More informationHuman and Optimal Exploration and Exploitation in Bandit Problems
Human and Optimal Exploration and ation in Bandit Problems Shunan Zhang (szhang@uci.edu) Michael D. Lee (mdlee@uci.edu) Miles Munro (mmunro@uci.edu) Department of Cognitive Sciences, 35 Social Sciences
More informationDecision-making deficit in chronic migraine patients with medication overuse
DOI 10.1007/s10072-012-1071-4 ORAL COMMUNICATIONS Decision-making deficit in chronic migraine patients with medication overuse B. Biagianti L. Grazzi O. Gambini S. Usai R. Muffatti S. Scarone G. Bussone
More informationExploring the Influence of Particle Filter Parameters on Order Effects in Causal Learning
Exploring the Influence of Particle Filter Parameters on Order Effects in Causal Learning Joshua T. Abbott (joshua.abbott@berkeley.edu) Thomas L. Griffiths (tom griffiths@berkeley.edu) Department of Psychology,
More information