A practical introduction to using the drift diffusion model of decision-making in cognitive psychology, neuroscience, and health sciences

Social threat avoidance depends on action-outcome predictability

Avoiding threatening individuals is pivotal for adaptation to our social environment. Yet, it remains unclear whether social threat avoidance is subtended by goal-directed processes, in addition to stimulus-response associations. To test this, we manipulated outcome predictability during spontaneous approach/avoidance decisions from avatars displaying angry facial expressions. Across three virtual reality experiments, we showed that participants avoided more often when they could predict the outcome of their actions, indicating goal-directed processes. However, above-chance avoidance rate when facing unpredictable outcomes suggested that stimulus-response associations also played a role. We identified two latent classes of participants: the "goal-directed class" showed above-chance avoidance only in the predictable condition, while the "stimulus-response class" showed no credible difference between conditions but had a higher overall avoidance rate. The goal-directed class exhibited greater cardiac deceleration in the predictable condition, associated with better value integration in decision-making. Computationally, this class had an increased drift-rate in the predictable condition, reflecting increased value estimation of threat avoidance. In contrast, the stimulus-response class showed higher responsiveness to threat, indicated by increased drift-rate for avoidance and increased muscular activity at response time. These results support the central role of goal-directed processes in social threat avoidance and reveal its physiological and computational correlates.

Emotion-Driven Moral Evaluation: A Mechanistic Study Based on the Drift-Diffusion Model

BACKGROUND

Moral evaluation is identified as the first stage in the theory of moral judgment, and academics believe that it may align with the social intuitionist model. This study aims to prove that the model's emotional dominance hypothesis applies to moral evaluation by presenting a computational decision-making model that mathematically formalizes this emotional dominance decision-making process. We also compared different types of valence evaluation tasks to test the emotional priority hypothesis.

METHODS

We used a convenience sampling method to randomly recruit 30 enrolled college students. The drift-diffusion model was employed to analyze reaction times for words with various emotional and moral valences Additionally, we designed different valence evaluation tasks based on the response relevance hypothesis and evaluated the processing order through reaction time comparisons.

RESULTS

The analysis revealed that the emotional mechanism of immoral evaluation differs from moral evaluation. An increase in emotional valence accelerates the speed of evidence accumulation (v) for moral evaluation (M = 1.21, 0.2% < 0 < 99.8%) but lowers decision caution (a) in immoral evaluation (M = -0.64, 96.1% < 0 < 3.9%). In contrast, moral valence does not have a significant influence on evaluation processes (v, M = -0.28, 72.1% < 0 < 27.9%; a, M = -0.32, 79.3% < 0 < 20.7%). Furthermore, We found no significant difference in reaction times between moral and immoral words in the emotional evaluation task (F(1,29) = 0.55, p = 0.464, partial η2 = 0.02), but a significant difference existed in the moral evaluation task (F(1,29) = 17.99, p < 0.001, partial η2 = 0.38), indicating that the tendency of relatively fast immoral evaluation in emotional evaluation tasks may be caused by emotional priority.

CONCLUSIONS

Our findings support the intuitive model's emotional dominance hypothesis and introduce a new emotional mechanism into moral evaluation. This study clarifies the distinct emotional processes in moral and immoral evaluations, fills a gap in the research on moral evaluation, and offers insights into human decision-making in moral contexts.

A behavioral dataset of predictive decisions given trends in information across adulthood

Making early and good predictions is a critical feature of decision making in domains such as investing and predicting the spread of diseases. Past literature indicates that people use recent and longer-term trends to extrapolate future outcomes. Nonetheless, less is known about what differentiates the strategies people use to make better predictions than others. Furthermore, factors underlying predictive judgments could be an important behavioral component in psychosocial research investigating manic-depression, anxiety, and age effects. Additionally, predictive judgments may be moderated based on the experience of living in areas with greater income inequality. To address these issues, we used investment tasks where participants had to predict future outcomes of their investments based on a trend in information. In the task, participants predicted how many tokens a gold mine would produce on the twelfth turn. On each turn, participants could ask for more information at a cost, or make a prediction about whether the gold mine would produce more or less than 100 tokens by the 12th turn. The trend was determined by function type (exponential and inverse exponential functions), whether the function was more linear or curved (growth factors), and good or bad outcomes (final values). This paradigm could help disentangle to what degree people use recent or longer-term information to inform their predictive judgments. We used Qualtrics to conduct this study. We also collected questionnaire data quantifying anxiety, impulsivity, risk attitudes, manic-depressive symptoms, and other psychosocial characteristics. The study was administered to adults with age ranges across the lifespan (N = 360; 225 male, 132 female; 3 nonbinary; mean age: 44.3 years; SD: 15.4 years, min: 18 years, max: 78 years). Additionally, we sampled across areas with high- and low-income inequality, thereby allowing researchers to investigate if value-based decisions are associated with participants' local communities. We outline potential ways to use and reuse this data, including exploring how individual differences are associated with predictive judgments.

Auditory decision-making deficits after permanent noise-induced hearing loss

Loud noise exposure is one of the leading causes of permanent hearing loss. Individuals with noise-induced hearing loss (NIHL) suffer from speech comprehension deficits and experience impairments to cognitive functions such as attention and decision-making. Here, we tested whether a specific sensory deficit, NIHL, can directly impair auditory cognitive function. Gerbils were trained to perform an auditory decision-making task that involves discriminating between slow and fast presentation rates of amplitude-modulated (AM) noise. Decision-making task performance was assessed across pre-versus post-NIHL sessions within the same gerbils. A single exposure session (2 hours) to loud broadband noise (120 dB SPL) produced permanent NIHL with elevated threshold shifts in auditory brainstem responses (ABRs). Following NIHL, decision-making task performance was tested at sensation levels comparable to those prior to noise exposure in all animals. Our findings demonstrate NIHL diminished perceptual acuity, reduced attentional focus, altered choice bias, and slowed down evidence accumulation speed. Finally, video-tracking analysis of motor behavior during task performance demonstrates that NIHL can impact sensory-guided decision-based motor execution. Together, these results suggest that NIHL impairs the sensory, cognitive, and motor factors that support auditory decision-making.

Exploring the spatial interference effects elicited by social and non-social targets: A conditional accuracy function approach

Recent studies employing the spatial interference paradigm reveal qualitative differences in congruency effects between gaze and arrow targets. Typically, arrows produce a standard congruency effect (SCE), with faster responses when target direction aligns with its location. Conversely, gaze targets often lead to a reversed congruency effect (RCE), where responses are slower in similar conditions. We explored this dissociation using the Conditional Accuracy Function (CAF) to assess accuracy across reaction time bins. Using a hierarchical linear mixed modelling approach to compare cropped eyes, and full faces as social stimuli, and arrows as non-social stimuli, we synthesized findings from 11 studies, which led to three distinct models. The results showed that with non-social targets, incongruent trials exhibited lower accuracy rates in the first bin than in subsequent bins, while congruent trials maintained stable accuracy throughout the distribution. Conversely, social targets revealed a dissociation within the fastest responses; alongside a general reduction in accuracy for both congruency conditions, congruent trials resulted in even lower accuracy rates than incongruent ones. These results suggest with gaze targets that additional information, perhaps social, in addition to the automatic capture by the irrelevant target location, is being processed during the earlier stages of processing.

Investigating the effects of context, visual working memory, and inhibitory control in hybrid visual search

Introduction

In real-life scenarios, individuals frequently engage in tasks that involve searching for one of the distinct items stored in memory. This combined process of visual search and memory search is known as hybrid search. To date, most hybrid search studies have been restricted to average observers looking for previously well-memorized targets in blank backgrounds.

Methods

We investigated the effects of context and the role of memory in hybrid search by modifying the task's memorization phase to occur in all-new single trials. In addition, we aimed to assess how individual differences in visual working memory capacity and inhibitory control influence performance during hybrid search. In an online experiment, 110 participants searched for potential targets in images with and without context. A change detection and go/no-go task were also performed to measure working memory capacity and inhibitory control, respectively.

Results

We show that, in target present trials, the main hallmarks of hybrid search remain present, with a linear relationship between reaction time and visual set size and a logarithmic relationship between reaction time and memory set size. These behavioral results can be reproduced by using a simple drift-diffusion model. Finally, working memory capacity did not predict most search performance measures. Inhibitory control, when relationships were significant, could account for only a small portion of the variability in the data.

Discussion

This study provides insights into the effects of context and individual differences on search efficiency and termination.

Aversive memories can be weakened during human sleep via the reactivation of positive interfering memories

Recollecting painful or traumatic experiences can be deeply troubling. Sleep may offer an opportunity to reduce such suffering. We developed a procedure to weaken older aversive memories by reactivating newer positive memories during sleep. Participants viewed 48 nonsense words each paired with a unique aversive image, followed by an overnight sleep. In the next evening, participants learned associations between half of the words and additional positive images, creating interference. During the following non-rapid-eye-movement sleep, auditory memory cues were unobtrusively delivered. Upon waking, presenting cues associated with both aversive and positive images during sleep, as opposed to not presenting cues, weakened aversive memory recall while increasing positive memory intrusions. Substantiating these memory benefits, computational modeling revealed that cueing facilitated evidence accumulation toward positive affect judgments. Moreover, cue-elicited theta brain rhythms during sleep predominantly predicted the recall of positive memories. A noninvasive sleep intervention can thus modify aversive recollection and affective responses.

The effect of anxiety and its interplay with social cues when perceiving aggressive behaviours

Contextual cues and emotional states carry expectations and biases that are used to attribute meaning to what we see. In addition, emotional states, such as anxiety, shape our visual systems, increasing overall, and particularly threat-related, sensitivity. It remains unclear, however, how anxiety interacts with additional cues when categorising sensory input. This is especially important in social scenarios where ambiguous gestures are commonplace, thus requiring the integration of cues for a proper interpretation. To this end, we decided to assess how states of anxiety might bias the perception of potentially aggressive social interactions, and how external cues are incorporated in this process. Participants (N = 71) were tasked with signalling the presence of aggression in ambiguous social interactions. Simultaneously, an observer (facial expression) reacted (by showing an emotional expression) to this interaction. Importantly, participants performed this task under safety and threat of shock conditions. Decision measures and eye-tracking data were collected. Our results showed that threat of shock did not affect sensitivity nor criterion when detecting aggressive interactions. The same pattern was observed for response times. Drift diffusion modelling analysis, however, suggested quicker evidence accumulation when under threat. Finally, dwell times over the observer were higher when under threat, indicating a possible association between anxiety states and a bias towards potentially threat-related indicators. Future probing into this topic remains a necessity to better explain the current findings.

Crowdsourcing with the drift diffusion model of decision making

Crowdsourcing involves the use of annotated labels with unknown reliability to estimate ground truth labels in datasets. A common task in crowdsourcing involves estimating reliabilities of annotators (such as through the sensitivities and specificities of annotators in the binary label setting). In the literature, beta or dirichlet distributions are typically imposed as priors on annotator reliability. In this study, we investigated the use of a neuroscientifically validated model of decision making, known as the drift-diffusion model, as a prior on the annotator labeling process. Two experiments were conducted on synthetically generated data with non-linear (sinusoidal) decision boundaries. Variational inference was used to predict ground truth labels and annotator related parameters. Our method performed similarly to a state-of-the-art technique (SVGPCR) in prediction of crowdsourced data labels and prediction through a crowdsourced-generated Gaussian process classifier. By relying on a neuroscientifically validated model of decision making to model annotator behavior, our technique opens the avenue of predicting neuroscientific biomarkers of annotators, expanding the scope of what may be learnt about annotators in crowdsourcing tasks.

Improved Perception of Aggression Under (un)Related Threat of Shock

Anxiety shifts visual attention and perceptual mechanisms, preparing oneself to detect potentially threatening information more rapidly. Despite being demonstrated for threat-related social stimuli, such as fearful expressions, it remains unexplored if these effects encompass other social cues of danger, such as aggressive gestures/actions. To this end, we recruited a total of 65 participants and asked them to identify, as quickly and accurately as possible, potentially aggressive actions depicted by an agent. By introducing and manipulating the occurrence of electric shocks, we induced safe and threatening conditions. In addition, the association between electric shocks and aggression was also manipulated. Our result showed that participants have improved sensitivity, with no changes to criterion, when detecting aggressive gestures during threat compared to safe conditions. Furthermore, drift diffusion model analysis showed that under threat participants exhibited faster evidence accumulation toward the correct perceptual decision. Lastly, the relationship between threat source and aggression appeared to not impact any of the effects described above. Overall, our results indicate that the benefits gained from states of anxiety, such as increased sensitivity toward threat and greater evidence accumulation, are transposable to social stimuli capable of signaling danger other than facial expressions.

Memory for rewards guides retrieval

Rewards paid out for successful retrieval motivate the formation of long-term memory. However, it has been argued that the Motivated Learning Task does not measure reward effects on memory strength but decision-making during retrieval. We report three large-scale online experiments in healthy participants (N = 200, N = 205, N = 187) that inform this debate. In experiment 1, we found that explicit stimulus-reward associations formed during encoding influence response strategies at retrieval. In experiment 2, reward affected memory strength and decision-making strategies. In experiment 3, reward affected decision-making strategies only. These data support a theoretical framework that assumes that promised rewards not only increase memory strength, but additionally lead to the formation of stimulus-reward associations that influence decisions at retrieval.

Impairment in recognition memory may be associated with near-term risk for suicide attempt in a high-risk sample

INTRODUCTION

Prior work has implicated several neurocognitive domains, including memory, in patients with a history of prior suicide attempt. The current study evaluated whether a delayed recognition test could enhance prospective prediction of near-term suicide outcomes in a sample of patients at high-risk for suicide.

METHODS

132 Veterans at high-risk for suicide completed a computer-based recognition memory test including semantically-related and -unrelated words. Outcomes were coded as actual suicide attempt (ASA), other suicide-related event (OtherSE) such as aborted/interrupted attempt or preparatory behavior, or neither (noSE), within 90 days after testing.

RESULTS

Reduced performance was a significant predictor of upcoming ASA, but not OtherSE, after controlling for standard clinical variables such as current suicidal ideation and history of prior suicide attempt. However, compared to the noSE reference group, the OtherSE group showed a reduction in the expected benefit of semantic relatedness in recognizing familiar words. A computational model, the drift diffusion model (DDM), to explore latent cognitive processes, revealed the OtherSE group had decreased decisional efficiency for semantically-related compared to semantically-unrelated familiar words.

LIMITATIONS

This study was a secondary analysis of an existing dataset, involving participants in a treatment trial, and requires replication; ~10 % of the sample was excluded from analysis due to failure to master the practice tasks and/or apparent noncompliance.

CONCLUSION

Impairments in recognition memory may be associated with near-term risk for suicide attempt, and may provide a tool to improve prediction of when at-risk individuals may be transitioning into a period of heightened risk for suicide attempt.

Similarities of SNARC, cognitive Simon, and visuomotor Simon effects in terms of response time distributions, hand-stimulus proximity, and temporal dynamics

The spatial-numerical association of response codes (SNARC) and Simon effects are attributed to the same type of conflict according to dimensional overlap (DO) theory: the congruency of task-irrelevant spatial information and the selected response (e.g., left or right). However, previous studies have yielded inconsistent results regarding the relationship between the two effects, with some studies reporting an interaction while others did not. This discrepancy may be attributed to the use of different types of Simon effects (visuomotor and cognitive Simon effects) in these studies, as the spatial codes associated with these two types of Simon effects are distinct (exogenous and endogenous, respectively). The aim of this study was to address these inconsistencies and gain a better understanding of the similarities and differences in spatial representations generated by spatial location, semantic information, and numerical information. We attempted to classify the relationships among the SNARC and Simon effects. Specifically, the visuomotor Simon, cognitive Simon, and SNARC effects were compared from three perspectives: the response time (RT) distribution, hand-stimulus proximity, and temporal dynamics (with the drift diffusion model; DDM). Regarding RTs, the results showed that the visuomotor Simon effect decreased with increased values of RT bins, while the cognitive Simon and SNARC effects increased. Additionally, the visuomotor Simon effect was the only effect influenced by hand-stimulus proximity, with a stronger effect observed in the hand-proximal condition than in the hand-distal condition. Regarding the DDM results, only the visuomotor Simon effect exhibited a higher drift rate and longer non-decision time in the incompatible condition than in the compatible condition. Conversely, both the SNARC and cognitive Simon effects exhibited an inverse pattern regarding the drift rate and no significant difference in non-decision time between the two conditions. These findings suggest that the SNARC effect is more similar to the cognitive Simon effect than the visuomotor Simon effect, indicating that the endogenous spatial-numerical representation of the SNARC effect might share an underlying processing mechanism with the endogenous spatial-semantic representation of the cognitive Simon effect but not with the exogenous location representation of the visuomotor Simon effect. Our results further demonstrate that the origin of spatial information could impact the classification of conflicts and supplement DO theory.

Listening to the Data: Computational Approaches to Addiction and Learning

Computational approaches hold great promise for identifying novel treatment targets and creating translational therapeutics for substance use disorders. From circuitries underlying decision-making to computationally derived neural markers of drug-cue reactivity, this review is a summary of the approaches to data presented at our 2023 Society for Neuroscience Mini-Symposium. Here, we highlight data- and hypothesis-driven computational approaches that recently afforded advancements in addiction and learning neuroscience. First, we discuss the value of hypothesis-driven algorithmic modeling approaches, which integrate behavioral, neural, and cognitive outputs to refine hypothesis testing. Then, we review the advantages of data-driven dimensionality reduction and machine learning methods for uncovering novel predictor variables and elucidating relationships in high-dimensional data. Overall, this review highlights recent breakthroughs in cognitive mapping, model-based analysis of behavior/risky decision-making, patterns of drug taking, relapse, and neuromarker discovery, and showcases the benefits of novel modeling techniques, across both preclinical and clinical data.

Evidence Accumulation Rate Moderates the Relationship between Enriched Environment Exposure and Age-Related Response Speed Declines

Older adults exposed to enriched environments (EEs) maintain relatively higher levels of cognitive function, even in the face of compromised markers of brain health. Response speed (RS) is often used as a simple proxy to measure the preservation of global cognitive function in older adults. However, it is unknown which specific selection, decision, and/or motor processes provide the most specific indices of neurocognitive health. Here, using a simple decision task with electroencephalography (EEG), we found that the efficiency with which an individual accumulates sensory evidence was a critical determinant of the extent to which RS was preserved in older adults (63% female, 37% male). Moreover, the mitigating influence of EE on age-related RS declines was most pronounced when evidence accumulation rates were shallowest. These results suggest that the phenomenon of cognitive reserve, whereby high EE individuals can better tolerate suboptimal brain health to facilitate the preservation of cognitive function, is not just applicable to neuroanatomical indicators of brain aging but can be observed in markers of neurophysiology. Our results suggest that EEG metrics of evidence accumulation may index neurocognitive vulnerability of the aging brain.Significance Statement Response speed in older adults is closely linked with trajectories of cognitive aging. Here, by recording brain activity while individuals perform a simple computer task, we identify a neural metric that is a critical determinant of response speed. Older adults exposed to greater cognitive and social stimulation throughout a lifetime could maintain faster responding, even when this neural metric was impaired. This work suggests EEG is a useful technique for interrogating how a lifetime of stimulation benefits brain health in aging.