Approach-Induced Biases in Human Information Sampling

Abstract

Information sampling is often biased towards seeking evidence that confirms one’s prior beliefs. Despite such biases being a pervasive feature of human behavior, their underlying causes remain unclear. Many accounts of these biases appeal to limitations of human hypothesis testing and cognition, de facto evoking notions of bounded rationality, but neglect more basic aspects of behavioral control. Here, we investigated a potential role for Pavlovian approach in biasing which information humans will choose to sample. We collected a large novel dataset from 32,445 human subjects, making over 3 million decisions, who played a gambling task designed to measure the latent causes and extent of information-sampling biases. We identified three novel approach-related biases, formalized by comparing subject behavior to a dynamic programming model of optimal information gathering. These biases reflected the amount of information sampled (“positive evidence approach”), the selection of which information to sample (“sampling the favorite”), and the interaction between information sampling and subsequent choices (“rejecting unsampled options”). The prevalence of all three biases was related to a Pavlovian approach-avoid parameter quantified within an entirely independent economic decision task. Our large dataset also revealed that individual differences in the amount of information gathered are a stable trait across multiple gameplays and can be related to demographic measures, including age and educational attainment. As well as revealing limitations in cognitive processing, our findings suggest information sampling biases reflect the expression of primitive, yet potentially ecologically adaptive, behavioral repertoires. One such behavior is sampling from options that will eventually be chosen, even when other sources of information are more pertinent for guiding future action.

Triple dissociation of attention and decision computations across prefrontal cortex

“…Recent debates on the role of PFC subregions in value-guided decision making 22 have included whether PFC acts in serial (certain subregions preceding others) or 23 parallel (simultaneous, distributed activity across subregions) 7,8 ; whether and how 24 attention shapes the decision process, in particular in OFC [9][10][11][12][13] ; and whether ACC 25 integrates evidence for different actions [14][15][16][17] , modifies behaviour in light of new 26 evidence [18][19][20] , or evaluates evidence for alternative courses of action 21,22 . 27 We addressed all of these questions by recording activity from ACC, DLPFC and 28 OFC (n=189, 134 and 183 single neurons respectively, see Extended Data Figure 1 and 29 Methods for details) during a decision task performed by two macaque monkeys (M. 30 mulatta).…”

“…27 We addressed all of these questions by recording activity from ACC, DLPFC and 28 OFC (n=189, 134 and 183 single neurons respectively, see Extended Data Figure 1 and 29 Methods for details) during a decision task performed by two macaque monkeys (M. 30 mulatta). Our task design (Figure 1a) mirrored behavioural studies examining 31 attention-guided information search during sequential, multi-attribute choice 23,24 . Each 32 option, presented on left and right sides of the screen, comprised two pre-learned 33 picture cues, representing different attributes, probability and magnitude of juice 34 reward.…”

Information seeking and simulation: Roles of attention in guiding a goal-directed behavior

“…These influences of gaze on the contextual cues can be interpreted as a pure manifestation of information sampling as reported in the previous studies [28][29][30] . However, the gaze behavior and information seeking can be interpreted as a reflection of simulation process 9,10,31,38 . Specifically, these results can indicate that the simulation of association sequence made it easier to choose the associated target in that sequence and difficult to choose the target not in the simulated sequence.…”

“…These two types of the influences on information seeking behavior are treated separately 1,4,5 . However, recent studies suggest that there are interactions between the two factors 4,9,10 . There can be some mechanism mediating the interaction between the informational and reward-related factors or there can be an intrinsic link between the two factors.…”

“…This simulation-based account of attention is consistent with the recent results reporting the interactions of information seeking and reward/valence aspect of the information 4 . One study reported that information seeking behavior of humans in card selection task was biased toward more valuable option 9 . Even in monkeys, information seeking behavior with gaze response showed the preference for the advance information of the options with higher rewarding outcome 10 .…”