Documentation Matters: Human-Centered AI System to Assist Data Science Code Documentation in Computational Notebooks

Computational notebooks allow data scientists to express their ideas through a combination of code and documentation. However, data scientists often pay attention only to the code, and neglect creating or updating their documentation during quick iterations. Inspired by human documentation practices learned from 80 highly-voted Kaggle notebooks, we design and implement Themisto, an automated documentation generation system to explore how human-centered AI systems can support human data scientists in the machine learning code documentation scenario. Themisto facilitates the creation of documentation via three approaches: a deep-learning-based approach to generate documentation for source code, a query-based approach to retrieve online API documentation for source code, and a user prompt approach to nudge users to write documentation. We evaluated Themisto in a within-subjects experiment with 24 data science practitioners, and found that automated documentation generation techniques reduced the time for writing documentation, reminded participants to document code they would have ignored, and improved participants’ satisfaction with their computational notebook.

Actively Open-Minded Thinking, Bullshit Receptivity, and Susceptibility to Framing: Evaluating the Dual-Process Account in North America and Bulgaria

The framing effect occurs when different presentations of the same problem lead to predictably different preferences. The dual-process framework of higher cognition assumes that the effect violates rational principles, but alternative accounts and recent evidence have contested this interpretation. Contributing to this debate, we tested the dual-process assumption by investigating associations between susceptibility to framing and the willingness and ability to think in line with rational norms, conceptualized as actively open-minded thinking and pseudo-profound bullshit receptivity. We conducted two online studies among North American (N = 259) and Bulgarian (N = 248) university students and administered several framing problems within subjects, presumably a necessary condition for the associations to appear. Confirmatory factor analyses showed that susceptibility to framing was associated with decreased actively open-minded thinking and increased bullshit receptivity in both sites. Exploratory multi-group analyses demonstrated partial strong invariance and showed that the findings generalize across both sites in terms of direction and partially in terms of magnitude. These results broadly support the dual-process account of the framing effect. Our study further contributes to adapting existing measures to a novel setting and expanding the findings across borders and populations.

Situational Peer Dynamics and Crime Decisions

Objectives To investigate how peer dynamics, specifically interpersonal conversations between a potential offender and a peer, contemporaneous with a crime opportunity, influence perceptions of sanction certainty and social costs. Methods Data are analyzed from randomized experiments and hypothetical vignettes embedded within a nationwide, online survey ( n = 1,275). Vignettes were presented for three distinct crime opportunities, drunk driving, fighting, and insurance fraud. Results The findings suggest that respondents adjust two core decision-making perceptions—the perceived certainty of being legally sanctioned and perceived social costs such as stigma or embarrassment—in accord with the content of verbal communications from peers. There is evidence for this both between and within subjects. Conclusions The study underscores the importance of accounting for both physical and social features of the situational context for crime in models of offender decision making. Implications are drawn regarding the social milieu for offender decision making, and the broader criminological relevance of choice principles.

Heterogeneity of social cognition in temporo-parietal junction: Overlapping yet distinct representation between visual perspective-taking and theory of mind

Visual perspective taking (VPT), particularly level 2 VPT (VPT2), which allows an individual to understand that the same object can be seen differently by others, is related to the theory of mind (ToM), because both functions require a decoupled representation from oneself. Although previous neuroimaging studies have shown that VPT and ToM activate the temporo-parietal junction (TPJ), it is unclear whether common neural substrates are involved in VPT and ToM. To clarify this point, the present study directly compared the TPJ activation patterns of individual participants performing VPT2 and ToM tasks using functional magnetic resonance imaging and within-subjects design. VPT2-induced activations were compared with activations observed during a mental rotation task as a control task, whereas ToM-related activities were identified with a standard ToM localizer using false-belief stories. A whole-brain analysis revealed that VPT2 and ToM activated overlapping areas in the posterior part of the TPJ. By comparing the activations induced by VPT2 and ToM in individual participants, we found that the peak voxels induced by ToM were located significantly more anteriorly and dorsally within the bilateral TPJ than those measured during the VPT2 task. We further confirmed that these activity areas were spatially distinct from the nearby extrastriate body area (EBA), visual motion area (MT+), and the posterior superior temporal sulcus (pSTS) using independent localizer scans. Our findings revealed that VPT2 and ToM have distinct representations, albeit partially overlapping, indicating the functional heterogeneity of social cognition within the TPJ.

Risk-Promoting Effects of Reward-Paired Cues in Human Sign- and Goal-Trackers

The incentive sensitization theory of addiction proposes that through repeated associations with addictive rewards, addiction-related stimuli acquire a disproportionately powerful motivational pull on behaviour. Animal research suggests trait-like individual variation in the degree of incentive salience attribution to reward-predictive cues, defined phenotypically as sign-tracking (high) and goal-tracking (low incentive salience attribution). While these phenotypes have been linked to addiction features in rodents, their translational validity has been little studied. Here, we examined whether sign- and goal-tracking in healthy human volunteers modulates the effects of reward-paired cues on cost-benefit decision making. Sign-tracking was measured in a Pavlovian conditioning paradigm as the amount of eye gaze fixation on the reward-predictive cue versus the location of impending reward delivery. In Study 1 (Cherkasova et al, 2018), participants were randomly assigned to perform a two-choice lottery task in which rewards were either accompanied (cued, n=63) or unaccompanied (uncued, n=68) by money images and casino jingles. In Study 2, participants (n=58) performed cued and uncued versions of the task in a within-subjects design. Across both studies, cues promoted riskier choice, and both studies yielded evidence of goal-tracking being associated with greater risk-promoting effects of cues. These findings are at odds with the notion of sign-trackers being preferentially susceptible to the influence of reward cues on behavior and point to the role of mechanisms besides incentive salience in mediating such influences.

Effect of Ambient Lighting on Frequency Dependence in Transcranial Electrical Stimulation-Induced Phosphenes

Inconsistencies have been found in the relationship between ambient lighting conditions and frequency-dependence in transcranial electric current stimulation (tECS) induced phosphenes. Using a within-subjects design across lighting condition (dark, mesopic [dim], photopic [bright]) and tECS stimulation frequency (10, 13, 16, 18, 20 Hz), this study determined phosphene detection thresholds in 24 subjects receiving tECS using an FPz-Cz montage. Minima phosphene thresholds were found at 16 Hz in mesopic, 10 Hz in dark and 20 Hz in photopic lighting conditions, with these thresholds being substantially lower for mesopic than both dark (60% reduction) and photopic (56% reduction), conditions. Further, whereas the phosphene threshold-stimulation frequency relation was linear in the dark (increasing with frequency) and photopic (decreasing with frequency) conditions, a quadratic function was found for the mesopic condition (where it followed the linear increase of the dark condition from 10-16 Hz, and the linear decrease of the photopic condition from 16-20 Hz). The results clearly demonstrate that ambient lighting is an important factor in the detection of tECS-induced phosphenes, and that mesopic conditions are most suitable for obtaining overall phosphene thresholds.

Disrupted reward processing in Parkinson’s disease and its relationship with dopamine state and neuropsychiatric syndromes: a systematic review and meta-analysis

BackgroundNeuropsychiatric symptoms are common in Parkinson’s disease (PD) and predict poorer outcomes. Reward processing dysfunction is a candidate mechanism for the development of psychiatric symptoms including depression and impulse control disorders (ICDs). We aimed to determine whether reward processing is impaired in PD and its relationship with neuropsychiatric syndromes and dopamine replacement therapy.MethodsThe Ovid MEDLINE/PubMed, Embase and PsycInfo databases were searched for articles published up to 5 November 2020. Studies reporting reward processing task performance by patients with PD and healthy controls were included. Summary statistics comparing reward processing between groups were converted to standardised mean difference (SMD) scores and meta-analysed using a random effects model.ResultsWe identified 55 studies containing 2578 participants (1638 PD and 940 healthy controls). Studies assessing three subcomponent categories of reward processing tasks were included: option valuation (n=12), reinforcement learning (n=37) and reward response vigour (n=6). Across all studies, patients with PD on medication exhibited a small-to-medium impairment versus healthy controls (SMD=0.34; 95% CI 0.14 to 0.53), with greater impairments observed off dopaminergic medication in within-subjects designs (SMD=0.43, 95% CI 0.29 to 0.57). Within-subjects subcomponent analysis revealed impaired processing off medication on option valuation (SMD=0.57, 95% CI 0.39 to 0.75) and reward response vigour (SMD=0.36, 95% CI 0.13 to 0.59) tasks. However, the opposite applied for reinforcement learning, which relative to healthy controls was impaired on-medication (SMD=0.45, 95% CI 0.25 to 0.65) but not off-medication (SMD=0.28, 95% CI −0.03 to 0.59). ICD was the only neuropsychiatric syndrome with sufficient studies (n=13) for meta-analysis, but no significant impairment was identified compared tonon-ICD patients (SMD=−0.02, 95% CI −0.43 to 0.39).ConclusionReward processing disruption in PD differs according to subcomponent and dopamine medication state, and warrants further study as a potential treatment target and mechanism underlying associated neuropsychiatric syndromes.

Losing sleep by staying up late leads adolescents to consume more carbohydrates and a higher glycemic load

This study examined how short sleep impacts dietary consumption in adolescents by testing whether experimentally shortening sleep influences the amount, macronutrient content, food types, and timing of food consumed. Ninety-three adolescents completed a within-subjects crossover paradigm comparing five nights of short sleep (6.5-hour sleep opportunity) to five nights of Healthy Sleep (9.5-hour sleep opportunity). Within each condition, adolescents completed three multiple-pass dietary recalls that recorded the types, amount, and timing of food intake. The following outcomes were averaged across days of dietary recall within condition: kilocalories, grams of carbohydrates, fat, protein, and added sugars, glycemic load of foods, and servings of specific types of foods (low-calorie drinks, sweetened drinks, fruits/vegetables, meats/proteins, processed snacks, “fast food” entrees, grains, and sweets/desserts). Timing of consumption of kilocalorie and macronutrient outcomes were also examined across four noncumulative time bins: 06:00–10:59, 11:00–15:59, 16:00–20:59, and 21:00–01:00. Adolescents slept 2 h and 20 min longer in Healthy Sleep than in Short Sleep (p < .0001). While in Short Sleep, adolescents ate more grams of carbohydrates (p = .031) and added sugars (p = .047), foods higher in glycemic load (p = .013), and servings of sweet drinks (p = .023) and ate fewer servings of fruits/vegetables (p = .006) compared to Healthy Sleep. Differences in consumption of kilocalories, fat, and carbohydrates emerged after 9:00 pm (ps = .012, .043, .006, respectively). These experimental findings suggest that adolescents who have insufficient sleep exhibit dietary patterns that may increase the risk for negative weight and cardiometabolic outcomes. Future health promotion efforts should include promoting optimal sleep to increase healthy dietary habits.

Effect of Trataka (Yogic Visual Concentration) on the Performance in the Corsi-Block Tapping Task: A Repeated Measures Study

Background and Objective: Attention and memory are essential aspects of cognitive health. Yogasanas, pranayama, and meditation have shown to improve cognitive functions. There has been no assessment of Trataka (yogic visual concentration) on working or on spatial memory. The present study was planned to assess the immediate effects of Trataka and of eye exercise sessions on the Corsi-block tapping task (CBTT).Methods: A total of 41 healthy volunteers of both genders with age 23.21 ± 2.81 years were recruited. All participants underwent baseline assessment, followed by 2 weeks of training in Trataka (including eye exercise). Each training session lasted for 20 min/day for 6 days a week. After completion of the training period, a 1-week washout period was given. Each participant then was assessed in two sessions in Trataka and in eye exercise on two separate days, maintaining the same time of the day. Repeated measure analysis of variance with Holm’s adjustment was performed to check the difference between the sessions.Results: Significant within-subjects effects were observed for forward Corsi span andforward total score (p < 0.001), and also for backward Corsi span (p < 0.05) and backward total score (p < 0.05). Post hoc analyses revealed Trataka session to be better than eye exercises and baseline. The eye exercise session did not show any significant changes in the CBTT.Conclusion: The result suggests that Trataka session improves working memory, spatial memory, and spatial attention.