Digital embryos: a novel technical approach to investigate perceptual categorization in pigeons (Columba livia) using machine learning

Pigeons are classic model animals to study perceptual category learning. To achieve a deeper understanding of the cognitive mechanisms of categorization, a careful consideration of the employed stimulus material and a thorough analysis of the choice behavior is mandatory. In the present study, we combined the use of “virtual phylogenesis”, an evolutionary algorithm to generate artificial yet naturalistic stimuli termed digital embryos and a machine learning approach on the pigeons’ pecking responses to gain insight into the underlying categorization strategies of the animals. In a forced-choice procedure, pigeons learned to categorize these stimuli and transferred their knowledge successfully to novel exemplars. We used peck tracking to identify where on the stimulus the animals pecked and further investigated whether this behavior was indicative of the pigeon’s choice. Going beyond the classical analysis of the binary choice, we were able to predict the presented stimulus class based on pecking location using a k-nearest neighbor classifier, indicating that pecks are related to features of interest. By analyzing error trials with this approach, we further identified potential strategies of the pigeons to discriminate between stimulus classes. These strategies remained stable during category transfer, but differed between individuals indicating that categorization learning is not limited to a single learning strategy.

A Satisficing Heuristic Decision-Making Model under Limited Attention and Incomplete Preferences

Classical choice theory assumes that a decision-maker considers all feasible alternatives. However, a decision-maker in the real world can not consider all alternatives because of limited attention. In this paper, we propose a satisficing choice model to describe the choice procedure based on the incomplete preferences under the limited attention of the decision-maker. Moreover, the existence and rationality properties of the satisficing choice model on the different domains are studied combined with some proposed rationality conditions. Further, the proposed satisficing choice model is applied to a case of quality competition. Results show that the satisficing choice model of this paper is of a certain theoretical guiding significance to a kind of emergency decisions made by decision-makers under the circumstance of time pressure and limited information. It can also be the theoretical foundation for the study on the boundedly rational decision-making.

Age- and frequency-dependent changes in dynamic contrast perception in visual snow syndrome

Objective Patients with visual snow syndrome (VSS) suffer from a debilitating continuous (“TV noise-like”) visual disturbance. They report problems with vision at night and palinopsia despite normal visual acuity. The underlying pathophysiology of VSS is largely unknown. Currently, it is a clinical diagnosis based on the patient’s history, an objective test is not available. Here, we tested the hypothesis that patients with VSS have an increased threshold for detecting visual contrasts at particular temporal frequencies by measuring dynamic contrast detection-thresholds. Methods Twenty patients with VSS were compared to age-, gender-, migraine- and aura-matched controls in this case-control study. Subjects were shown bars randomly tilted to the left or right, flickering at six different frequencies (15 Hz, 20 Hz, 25 Hz, 30 Hz, 35 Hz, 40 Hz). The contrast threshold (CT) for detection of left or right tilt was measured in a two-alternative adaptive forced-choice procedure (QUEST). The threshold was defined as the Michelson contrast necessary to achieve the correct response in 75% of the cases. Results The CT increased for higher flicker frequencies (ANOVA: main effect frequency: F (5,180) = 942; p < 0.001), with an additional significant frequency*diagnosis interaction (ANOVA: F (5,180) = 5.00; p < 0.001). This interaction effect was due to an increased CT at a flicker frequency of 15 Hz in the VSS cohort (VSS: MC = 1.17%; controls: MC = 0.77%). At the other frequencies, group comparisons revealed no differences. Furthermore, in the VSS cohort we observed an increase of CT with higher age (r = 0.69; p < 0.001), which was not seen in controls (r = 0.30; p = 0.20). Conclusions This study demonstrates a lower visual contrast sensitivity exclusively at 15 Hz in VSS patients and demonstrates frequency-dependent differences in dynamic contrast vision. The peak sensitivities of both parvo- and magnocellular visual pathways are close to a frequency of about 10 Hz. Therefore, this frequency seems to be of crucial importance in everyday life. Thus, it seems plausible that the impairment of contrast sensitivity at 15 Hz might be an important pathophysiological correlate of VSS. Furthermore, the overall age-related decrease in contrast sensitivity only in VSS patients underscores the vulnerability of dynamic contrast detection in VSS patients. Dynamic CT detection seems to be a promising neurophysiological test that may contribute to the diagnosis of VSS.

The effect of blue light filtering lenses on speed perception

Blue-light filtering lenses (BFLs) are marketed to protect the eyes from blue light that may be hazardous to the visual system. Because BFLs attenuate light, they reduce object contrast, which may impact visual behaviours such as the perception of object speed which reduces with contrast. In the present study, we investigated whether speed perception is affected by BFLs. Using a two-interval forced-choice procedure in conjunction with Method of Constant Stimuli, participants (n = 20) judged whether the perceived speed of a moving test stimulus (1.5–4.5°/s) viewed through a BFL was faster than a reference stimulus (2.75°/s) viewed through a clear lens. This procedure was repeated for 3 different BFL brands and chromatic and achromatic stimuli. Psychometric function fits provided an estimate of the speed at which both test and reference stimuli were matched. We find that the perceived speed of both chromatic and achromatic test stimuli was reduced by 6 to 20% when viewed through BFLs, and lenses that attenuated the most blue-light produced the largest reductions in perceived speed. Our findings indicate that BFLs whilst may reduce exposure to hazardous blue light, have unintended consequences to important visual behaviours such as motion perception.

Towards a common code for difficulty: Navigating a narrow gap is like memorizing an extra digit

What makes a task hard or easy? The question seems easy, but answering it has been hard. The only consensus has been that, all else being equal, easy tasks can be performed by more individuals than hard tasks, and easy tasks are usually preferred over hard tasks. Feghhi and Rosenbaum (Journal of Experimental Psychology: Human Perception and Performance, 45, 983–994, 2019) asked whether task difficulty might reflect a single amodal quantity. Based on their subjects’ two-alternative forced-choice data from tasks involving choices of tasks with graded physical and mental challenges, the authors showed that the difficulty of passing through a narrow gap rather than a wide gap was psychologically equivalent to memorizing an extra .55 digits. In the present study, we extended this approach by adding new arguments for the hypothesis that task difficulty might reflect a single amodal quantity (inspired by considerations of physics, economics, and the common code hypothesis for the study of perception and action), and we tested narrower gaps than before to see whether we would find a larger equivalent memory-digit. Consistent with our prediction, we obtained a value of .95. We suggest that our multi-modal two-alternative forced-choice procedure can pave the way toward a better understanding of task difficulty.

Individual difference in addiction-like behaviors and choice between cocaine versus food in Heterogeneous Stock rats

Rationale and objectives: Recent studies reported that when given a mutually exclusive choice between cocaine and palatable food, most rats prefer the non-drug reward over cocaine. However, these studies used rat strains with limited genetic and behavioral diversity. Here, we used a unique outbred strain of rats (Heterogeneous Stock, HS) that mimic the genetic variability of humans. Methods: We first identified individual differences in addiction-like behaviors (low and high). Next, we tested choice between cocaine and palatable food using a discrete choice procedure. We characterized the individual differences using an Addiction score that incorporates key features of addiction: escalated intake, highly motivated responding (progressive ratio), and responding despite adverse consequences (footshock punishment). We assessed food vs. cocaine choice at different drug-free days (without pre-trial cocaine self administration) during acquisition of cocaine self-administration or after escalation of cocaine self-administration. We also assessed drug vs. food choice immediately after 1-, 2-, or 6-h cocaine self-administration. Results: Independent of the addiction score, without pre-trial coccaine (1 or more abstinence days) HS rats strongly preferred the palatable food over cocaine, even if the food reward was delayed or its size was reduced. However, rats with high but not low addiction score modestly increased cocaine choice immediately after 1-, 2- or 6-h cocaine self-administration. Conclusions: Like other strains, HS rats strongly prefer palatable food over cocaine. Individual differences in addiction score were associated with increased drug choice in the presence but not absence (abstinence) of cocaine. The HS strain may be useful in studies on mechanisms of addiction vulnerability.

Safety and Preliminary Efficacy of Lorcaserin for Cocaine Use Disorder: A Phase I Randomized Clinical Trial

Background and Objectives: Preclinical studies show serotonin (5-HT) 5-HT2C receptor (5-HT2CR) agonists reduce cocaine-seeking and cocaine intake. This study examined safety of the 5-HT2CR agonist lorcaserin administered with cocaine in participants with cocaine use disorder (CocUD). Secondarily, subjective response to cocaine and choice of cocaine vs. money were examined.Methods: A double-blind, randomized, placebo-controlled trial of 25 inpatient non-treatment seeking participants with CocUD. Participants were randomized to either lorcaserin (n = 17) or placebo (n = 8). Primary outcome measures included cardiovascular measures and plasma cocaine levels. Secondary measures of subjective response to cocaine were assessed using a visual analog scale (VAS) and cocaine vs. money progressive ratio choice sessions.Results: Thirteen randomized participants were included in the final analysis. No serious or unexpected adverse events were related to lorcaserin. There were no significant interactions between cocaine and lorcaserin on cardiovascular measures, plasma cocaine, or subjective ratings. After multiple comparisons correction, cocaine significantly increased blood pressure, heart rate, and QTc. Lorcaserin significantly decreased VAS ratings of “feel irritable,” “feel hungry,” and “I am craving.” For the cocaine vs. money choice procedure, there was a significant interaction between choice (cocaine vs. money) and lorcaserin. Participants treated with lorcaserin were more likely to choose cocaine.Discussion and Conclusions: This study showed safety of lorcaserin administered with cocaine but lack of efficacy to reduce the reinforcing effects of cocaine.Scientific Significance: This study is the first to show a disconnect between effects of 5-HT2CR agonists on craving and cocaine choice in human cocaine users.

Self-Control Preferences and Status-Quo Bias

In a setting of choice with an observable status quo, we model an agent who struggles with temptation by exercising (costly) self-control, and who views the status quo as a commitment opportunity that allows him to avoid the self-control costs incurred when making an active choice. Our model is rational in that the agent always maximizes the same ex-post utility function; hence, when the standard indirect utility property holds, the model reduces to classic rational choice model. However, when we allow for costly self-control, our theory provides a rationale for three well-documented phenomena that cannot be captured by the standard model: the status-quo bias, the compromise effect, and the satisficing choice procedure (when ignoring the information on the status quo). A notable feature of our theory is that while it relaxes the indirect utility property, it still allows for a complete identification of an agent’s preference relation over menus. This is found to be important in many practical situations.