Contrast effects or loss aversion? Comment on Usher and McClelland (2004)

M. Usher and J. L. McClelland (2004) recently proposed a new connectionist type of model to explain context effects on preferential choice including the similarity, attraction, and compromise effects. They compared their model with an earlier connectionist type model for these same effects proposed by R. Roe, J. R. Busemeyer, and J. T. Townsend (2001) and raised several new issues. The authors address these issues and point out the main theoretical differences between the 2 explanations for context effects.     

Preference reversal in multiattribute choice

A central puzzle for theories of choice is that people's preferences between options can be reversed by the presence of decoy options (that are not chosen) or by the presence of other irrelevant options added to the choice set. Three types of reversal effect reported in the decision-making literature, the attraction, compromise, and similarity effects, have been explained by a number of theoretical proposals. Yet a major theoretical challenge is capturing all 3 effects simultaneously. We review the range of mechanisms that have been proposed to account for decoy effects and analyze in detail 2 computational models, decision field theory (Roe, Busemeyer, & Townsend, 2001) and leaky competing accumulators (Usher & McClelland, 2004), that aim to combine several such mechanisms into an integrated account. By simulating the models, we examine differences in the ways the decoy effects are predicted. We argue that the LCA framework, which follows on Tversky's relational evaluation with loss aversion (Tversky & Kahneman, 1991), provides a more robust account, suggesting that common mechanisms are involved in both high-level decision making and perceptual choice, for which LCA was originally developed.     

On loss aversion in capuchin monkeys

Chen, Lakshminarayanan, and Santos (2006) claim to show in three choice experiments that monkeys react rationally to price and wealth shocks, but, when faced with gambles, display hallmark, human-like biases that include loss aversion. We present three experiments with monkeys and humans consistent with a reinterpretation of their data that attributes their results not to loss aversion, but to differences between choice alternatives in delay of reinforcement.     

Stimulus generalization of conditioned taste aversion in rats

Relatively little information is available regarding the intradimensional stimulus generalization of conditioned taste aversion (CTA). Experiment 1 employed a between-groups generalization test to examine the extent to which conditioned flavor aversion to one sucrose solution generalized to other concentrations of sucrose in adult rats. Evidence of a gradient of aversion was obtained. Because generalization gradients in other tasks have been found to flatten over a retention interval, Experiment 2 investigated the effects of delayed testing (2, 7, or 21 days) upon the slope of the generalization gradient. The generalization gradient flattened at the intervals, suggesting that subjects forgot the specific attributes of the conditioning concentration and avoided generalized stimuli as if they were the original CS. Experiment 3 used a long delay between taste and toxicosis to degrade the associative contingency and found no evidence that the generalization gradients found in the first two experiments could be explained in terms of enhanced neophobia due to poisoning. These findings provide further evidence (cf. A. W. Logue, 1979, Psychological Bulletin, 86, 276-296; M. Domjan, 1980, in J. S. Rosenblatt, R. A. Hinde, C. Beer, & M. Busnel (Eds.), Advances in the study of behavior, Vol. 11, New York: Academic Press) that CTA shares a number of similarities with other learning processes. Further, they illustrate that stimulus forgetting can be detected in a paradigm considered relatively immune to retention loss.     

Inhibitory effects of repeating color and shape: inhibition of return or repetition blindness?

There is a current debate regarding whether attention is influenced by stimulus attributes other than location. The present article replicates and extends previous findings that repeating the nonspatial attribute of color leads to a delay in target detection (M. B. Law, J. Pratt, & R. A. Abrams, 1995). Repetition disadvantage effects were found for the stimulus attributes of both color and shape, as well as for location. However, the nonspatial repetition disadvantage disappeared if the stimuli were presented in peripheral locations (Experiments 3a, 3b, and 4) or the cue was presented for 50 ms (Experiment 6). Moreover, the magnitude of the repetition disadvantage tended to decline as the cue-target stimulus-onset asynchrony increased (Experiments 5a, 5b, and 6). These results suggest that a repetition blindness mechanism may underlie the repetition disadvantage effects of nonspatial features, rather than an inhibition of return mechanism.     

Multialternative decision field theory: a dynamic connectionist model of decision making

The authors interpret decision field theory (J. R. Busemeyer & J. T. Townsend, 1993) as a connectionist network and extend it to accommodate multialternative preferential choice situations. This article shows that the classic weighted additive utility model (see R. L. Keeney & H. Raiffa, 1976) and the classic Thurstone preferential choice model (see L. L. Thurstone, 1959) are special cases of this new multialternative decision field theory (MDFT), which also can emulate the search process of the popular elimination by aspects (EBA) model (see A. Tversky, 1969). The new theory is unique in its ability to explain several central empirical results found in the multialternative preference literature with a common set of principles. These empirical results include the similarity effect, the attraction effect, and the compromise effect, and the complex interactions among these three effects. The dynamic nature of the model also implies strong testable predictions concerning the moderating effect of time pressure on these three effects.     

Linking inhibition to activation in the control of task sequences

Inhibition of abandoned tasks in task switching can be inferred when a worse performance is found withn− 2 task repetitions (ABA sequences) than with nonrepetitions (CBA sequences). Recent evidence has shown that this inhibition effect decreases with long intertrial intervals (i.e., response-cue intervals, RCIs). Two alternatives have been proposed to account for this decrease. One alternative attributes the observed decrease to the decay of inhibition itself. The other alternative proposes that decay of the activation of competing tasks reduces the interference and leads to less inhibition. To decide between these alternatives, we manipulated RCI trialwise. The results favor the decay-of-activation account as an explanation for the decreased inhibition effect. This links the amount of inhibition to the activation level of the competing tasks, whereas evidence for the decay of inhibition remains weak.     

Loss-aversion or loss-attention: The impact of losses on cognitive performance

Losses were found to improve cognitive performance, and this has been commonly explained by increased weighting of losses compared to gains (i.e., loss aversion). We examine whether effects of losses on performance could be modulated by two alternative processes: an attentional effect leading to increased sensitivity to task incentives; and a contrast-related effect. Empirical data from five studies show that losses improve performance even when the enhanced performance runs counter to the predictions of loss aversion. In Study 1–3 we show that in various settings, when an advantageous option produces large gains and small losses, participants select this alternative at a higher rate than when it does not produce losses. Consistent with the joint influence of attention and contrast-related processes, this effect is smaller when a disadvantageous alternative produces the losses. In Studies 4 and 5 we find a positive effect on performance even with no contrast effects (when a similar loss is added to all alternatives). These findings indicate that both attention and contrast-based processes are implicated in the effect of losses on performance, and that a positive effect of losses on performance is not tantamount to loss aversion.     

MDFT account of decision making under time pressure

In this paper, decision making under time pressure for multiattribute choice alternatives in a risky environment is investigated. A model, multiattribute decision field theory (MDFT), is introduced that describes both the dynamic and the stochastic nature of decision making and accounts for the observed changes in choice probabilities, including preference reversals as a function of time limit. An experiment in which five different time limits were imposed on the decision maker is presented to test the predictions of the model. It is shown that MDFT is able to account for the complex decision behavior observed in the data. Furthermore, MDFT is compared with the predictions of decision field theory (Busemeyer & Townsend, 1993; Roe, Busemeyer, & Townsend, 2001).     

Does auditory streaming require attention? Evidence from attentional selectivity in short-term memory

R. P. Carlyon, R. Cusack, J. M. Foxton, and I. H. Robertson (2001) have argued that attention is crucial for auditory streaming. The authors review R. P. Carlyon et al.'s (2001) arguments and suggest that a pertinent literature, the irrelevant sound paradigm--demonstrating preattentive auditory streaming--has been overlooked. In illustration of this alternative approach, the authors include a novel single experiment demonstrating the impact of preattentive auditory streaming on short-term serial memory. It is concluded that R. P. Carlyon et al.'s (2001) results do not definitively demonstrate that auditory streaming processes are dependent on attention; indeed, they are compatible with alternative accounts of the relationship between perceptual organization and attention.     

Attention to novelty, fear-anxiety, and age: their effects on conduct problems

The authors investigated the effects of attention to novelty, fear-anxiety, and age on 3 measures of conduct problems. They found several main and interaction effects. The results indicated the presence of the hypothesized 3-way interaction for 2 dependent measures (i.e., conduct problem, socialized aggression); the 3rd dependent measure (i.e., conduct disorder) approached significance (p = .07). Participants who were older, had low attention to novelty, and high fear-anxiety generally had the highest conduct-problem scores among 8 comparison groups. Older students with high attention to novelty and low fear-anxiety had significantly lower conduct-problem scores that were unremarkable. The findings are congruent with J. Gray's (1987) theory of behavioral activation and inhibition, and the results support R. C. Eaves' (1993) integrated theory of human behavior, which postulates that the level of attention to novelty, fear-anxiety, and age interact to produce individuals with conduct problems.     

Conflict-mediated choice

Research on behavioral decision making has demonstrated that preferences are affected by the set of options under consideration. Formal models offer an account of such choice-set effects on the basis of principles according to which the similarity of the options in the choice set has an effect on choice probabilities. Behavioral formulations are Tversky's (1972) elimination-by-aspects (EBA) model, based on the principle of sequential elimination, and Tversky and Simonson's (1993) componential context (CC) model, based on the principle of pairwise comparisons subject to loss aversion. While these principles are assumed to determine choice behavior, the conflict induced by the set of options under consideration is assumed to be merely a source of indeterminacy in choice behavior. This paper challenges past models by offering a behavioral account of choice-set effects according to which the effect of similarity on choice probabilities is mediated by conflict. A conflict-mediated choice (CMC) model is developed, its testable implications are derived, and its predictions are contrasted with those of the EBA model and the CC model. Specifically, it is described how, according to each of the three models, choice probabilities are affected by the similarity of the options in the choice set and how choice-set effects are moderated by the relative weight of the attributes along which the options are positioned. In two experimental studies, the CMC model offers an accurate account of observed choice-set effects and outperforms the alternative models.     

Testing alternative explanations of phantom decoy effects

Phantom decoys are alternatives that asymmetrically dominate a targeted alternative and yet lead to increased selection of the target when the decoy is declared to be unavailable. This effect is difficult to explain within most standard theoretical accounts of decoy effects. The current experiments tested between three explanations of this effect: (1) the relative advantage model based on loss aversion, (2) similarity substitution, and (3) range weighting. In Experiment 1, participants were presented trinary choice sets, with half of the sets containing a phantom decoy in one of five possible locations within the attribute space. Phantom decoy effects were robust across all decoy locations but one, and the pattern of effects most closely corresponded to predictions of the relative advantage model. Experiment 2 used a within‐subjects manipulation of the five phantom decoy locations. The overall pattern of effects most closely corresponded to predictions from the relative advantage model, as did the pattern for the group of participants who exhibited the strongest phantom decoy effects. Copyright © 2007 John Wiley & Sons, Ltd.     

Can attentional theory explain the inverse base rate effect? Comment on Kruschke (2001)

In J. K. Kruschke's (2001; see record 2001-18940-005) study, it is argued that attentional theory is the sole satisfactory explanation of the inverse base rate effect and that eliminative inference (P. Juslin, P. Wennerholm, & A. Winman, 2001; see record 2001-07828-016) plays no role in the phenomenon. In this comment, the authors demonstrate that, in contrast to the central tenets of attentional theory, (a) rapid attention shifts as implemented in ADIT decelerate learning in the inverse base-rate task and (b) the claim that the inverse base-rate effect is directly caused by an attentional asymmetry is refuted by data. It is proposed that a complete account of the inverse base-rate effect needs to integrate attention effects with inference rules that are flexibly used for both induction and elimination.     

A ballistic model of choice response time

Almost all models of response time (RT) use a stochastic accumulation process. To account for the benchmark RT phenomena, researchers have found it necessary to include between-trial variability in the starting point and/or the rate of accumulation, both in linear (R. Ratcliff & J. N. Rouder, 1998) and nonlinear (M. Usher & J. L. McClelland, 2001) models. The authors show that a ballistic (deterministic within-trial) model using a simplified version of M. Usher and J. L. McClelland's (2001) nonlinear accumulation process with between-trial variability in accumulation rate and starting point is capable of accounting for the benchmark behavioral phenomena. The authors successfully fit their model to R. Ratcliff and J. N. Rouder's (1998) data, which exhibit many of the benchmark phenomena.     

On adaptation, maximization, and reinforcement learning among cognitive strategies

Analysis of binary choice behavior in iterated tasks with immediate feedback reveals robust deviations from maximization that can be described as indications of 3 effects: (a) a payoff variability effect, in which high payoff variability seems to move choice behavior toward random choice; (b) underweighting of rare events, in which alternatives that yield the best payoffs most of the time are attractive even when they are associated with a lower expected return; and (c) loss aversion, in which alternatives that minimize the probability of losses can be more attractive than those that maximize expected payoffs. The results are closer to probability matching than to maximization. Best approximation is provided with a model of reinforcement learning among cognitive strategies (RELACS). This model captures the 3 deviations, the learning curves, and the effect of information on uncertainty avoidance. It outperforms other models in fitting the data and in predicting behavior in other experiments.     

Adaptation-level theory, opponent process theory, and dispositions: an integrated approach to the stability of job satisfaction

Research suggests that the stability of job satisfaction is partially the result of dispositions (J. J. Connolly & C. Viswesvaran, 2000; C. Dormann & D. Zapf, 2001; T. A. Judge & J. E. Bono, 2001a; T. A. Judge, D. Heller, & M. K. Mount, 2002). Opponent process theory (R. L. Solomon & J. D. Corbit, 1973, 1974) and adaptation-level theory (H. Helson, 1948) are alternative explanations of this stability that explain how environmental effects on job satisfaction dissipate across time. On the basis of an integration of these explanations, the authors propose that dispositions (a) influence employees' equilibrium or adaptation level of job satisfaction, (b) influence employees' sensitivity to workplace events, and (c) influence the speed at which job satisfaction returns to equilibrium after one is exposed to a workplace event. Research and applied implications are discussed.     

Elemental representations of stimuli in associative learning

This article reviews evidence and theories concerning the nature of stimulus representations in Pavlovian conditioning. It focuses on the elemental approach developed in stimulus sampling theory (R. C. Atkinson & W. K. Estes, 1963; R. R. Bush & F. Mosteller, 1951b) and extended by I. P. L. McLaren and N. J. Mackintosh (2000, 2002) and contrasts this with models that invoke notions of configural representations that uniquely code for different patterns of stimulus inputs (e.g., J. M. Pearce, 1987, 1994; R. A. Rescorla & A. R. Wagner, 1972; A. R. Wagner & S. E. Brandon, 2001). The article then presents a new elemental model that emphasizes interactions between stimulus elements. This model is shown to explain a range of behavioral findings, including those (e.g., negative patterning and biconditional discriminations) traditionally thought to be beyond the explanatory capabilities of elemental models. Moreover, the model offers a ready explanation for recent findings reported by R. A. Rescorla (2000, 2001, 2002b) concerning the way that stimuli with different conditioning histories acquire associative strength when conditioned in compound.     

Executive control of visual attention in dual-task situations

A theory of executive control is presented that proposes that executive processes control subordinate processes by manipulating their parameters, reconfiguring them to respond in accord with the current task set. It adopts C. Bundesen's (1990) theory of visual attention (TVA) and R. M. Nosofsky and T. J. Palmeri's (1997) exemplar-based random walk (EBRW) as the theory of subordinate processes. It assumes that a task set is a set of TVA and EBRW parameters sufficient to perform a task and that set switching involves changing those parameters. The theory solves 2 computational problems that emerge in dual-task situations: the binding problem and the serial order problem. It can perform dual tasks in series or in parallel but prefers the serial strategy because it is faster and it solves the binding problem naturally. The theory accounts for concurrence cost, set-switching cost, crosstalk between tasks, and the modulation of crosstalk by task set.     

List composition and the word-frequency effect for recognition memory

The attention/likelihood theory (ALT; M. Glanzer & J. K. Adams, 1990) and the retrieving effectively from memory (REM) theory (R. M. Shiffrin & M. Steyvers, 1997) make different predictions concerning the effect of list composition on word recognition. The predictions were empirically tested for two-alternative forced-choice, yes-no, and ratings recognition tasks. In the current article, the authors found that discrimination of low-frequency words increased as the proportion of high-frequency words studied increased. The results disconfirm the ALT prediction that recognition is insensitive to list composition, and they disconfirm the predictions of the REM model described by R. M. Shiffrin and M. Steyvers (1997). The current authors discuss a slightly modified version of REM that can better predict our findings, and we discuss the challenges the present findings pose for ALT and REM.