Hi, I am Tianwei Gong.

Most research into causal learning has focused on atemporal contingency data settings while fewer studies have examined learning and reasoning about systems exhibiting events that unfold in continuous time. Of these, none have yet explored learning about preventative causal influences. We explore human causal structure learning within a space of hypotheses that combine generative and preventative causal relationships. We find that participants are capable learners in this setting. We lay out a computational-level framework for normative inference in this setting and propose a family of more cognitively plausible algorithmic approximations. We find that participants' judgment patterns can be both qualitatively and quantitatively captured by a model that approximates normative inference via a simulation and summary statistics scheme based on structurally local computation using temporally local evidence.

How do people learn the latent causal structure that connects events and actions as they unfold in continuous time? We examine how people actively learn about causal structure in a continuous-time setting, focusing on when and where they intervene and how this shapes their learning. Across two experiments, we find that participants’ accuracy depends on both the informativeness and evidential complexity of the data they generate. Moreover, participants’ intervention choices strike a balance between maximizing expected information and minimizing inferential complexity. People time and target their interventions to create simple yet informative causal dynamics. We discuss how the continuous-time setting challenges existing computational accounts of active causal learning, and argue that metacognitive awareness of one’s inferential limitations plays a critical role for successful learning in the wild.

(Impact Statement) Everyday decision making is shaped by our judgments about how the world works, such as whether a new vaccine is beneficial or harmful for our health. Since more evidence is arriving all the time, we inevitably have to make these judgments with incomplete information. Here we show for the first time that people may use the timing of what has already happened to predict what will happen next, and then incorporate this predicted future evidence into their causal judgments. For instance, we find settings in which people take a rising trend to indicate that greater problems are to come, or a falling trend to indicate that the peak of a causal influence has already passed, leading them to make dramatically different causal judgments from the same overall case statistics. These results have implications both for understanding the sophistication of individual human causal reasoning and for pinpointing the source of disagreements in public discourse around scientific evidence.

Spatial neglect has been a phenomenon of interest for perceptual and neuropsychological researchers for decades. However, the underlying cognitive processes remain unclear. We provide a Bayesian framework for the classic line bisection task in spatial neglect, regard- ing as rational inferences in the face of uncertain infor- mation. A Bayesian observer perceives the left and right endpoints of a line with uncertainty, and leverages prior expectations about line lengths to compensate for this uncertainty. This Bayesian model provides a basis for characterizing different patterns of behavior. Our model also captures the paradoxical cross-over effect observed in earlier studies. It provides measures that correlate well with measures from other neglect tests, and can accu- rately distinguish stroke patients from healthy controls.

Many of the real world phenomena that cognizers must grapple with are continuous, not only in the values they can take, but also in how these values change over time. The mind must somehow abstract from these inputs to extract useful discrete concepts such as objects, events and causal relationships. We investigate several factors that affect basic inferences about causal relationships between continuous variables based on observations in continuous time. In a novel experiment, we explore the ways in which causal judgments are sensitive to factors that relate to causal inductive biases (e.g. causal lags, the direction of variation) and causal perception (e.g. the range and rapidity of variation). We argue standard statistical time-series models have limited utility in accounting for human sensitivity to these factors. We suggest further work is needed to fully understand the cognitive processes that underlie causal induction from time-series information.

Self-paced reading paradigm has been popular in psycholinguistic research for several decades. FAB (named from Forward-And-Backward, but also, FABulous), is a dummy’s tool that helps researchers easily design, implement, and analyze their self-paced reading tasks. Its basis on web languages promotes experiment implement and material sharing in this open science era, while its form-based user interface does not require researchers to have any knowledge in programming languages. In addition, FAB has a unique forward-and-backward mode which can acquire the regressive-like behavior that is usually only be recorded by eye-tracking equipment. Try it if you are in psycholinguistics.

Wait a minute. FAB also support other experiments, since it can display stimuli (e.g., words, sentences, pictures, icons, mathematical formula, memorized items) and record the reaction time:) Use it to save your time for programming.

Course selection is an important issue for university students. This study investigates the relationship between course selection and overall academic performance among university students. Using anonymous transcripts from 1681 undergraduates across ten majors, we examined the association between elective course load or choices and academic performance while controlling for the influence of grade leniency. Results showed that students with higher academic performance tended to schedule more courses at the early stage of university, even when the average workload was already high. Additionally, students with different levels of academic performance showed differences in their elective course choices, with course choices positively associated with academic performance being rated as having a higher potential to satisfy extrinsic motivation but a lower potential to satisfy intrinsic motivation. We discussed the possibility of interpreting the association between course selection and academic performance from a psychological perspective.

Cognitive reflection refers to the tendency to override a prepotent intuitive response and to engage in further reflection that leads to the correct response. The present study aimed to validate a child-friendly version of the cognitive reflection test, the CRT-D, in China. We administered the CRT-D to 130 adults and 111 school-age children in China and compared performance on the CRT-D to several measures of rational thinking (belief bias syllogisms, base rate sensitivity, denominator neglect, otherside thinking) and normative thinking dispositions (actively open-minded thinking, need for cognition). The CRT-D was a significant predictor of rational thinking and normative thinking dispositions in both children and adults as previously found in the Western culture. Adults’ performance on the CRT-D correlated with their performance on the original CRT, and children’s performance on the CRT-D predicted rational thinking and normative thinking dispositions even after adjusting for age. These results demonstrate that the CRT-D has broad utility and that cognitive reflection, rational thinking, and normative thinking dispositions converge even in a non- Western culture.

Events that violate the laws of nature are, by definition, impossible, but recent research suggests that people view some violations as “more impossible” than others (Shtulman & Morgan, 2017). When evaluating the difficulty of magic spells, American adults are influenced by causal considerations that should be irrelevant given the spell’s primary causal violation, judging, for instance, that it would be more difficult to levitate a bowling ball than a basketball even though weight should no longer be a consideration if contact is no longer necessary for support. In the present study, we sought to test the generalizability of these effects in a non-Western context—China—where magical events are represented differently in popular fiction and where reasoning styles are often more holistic than analytic. Across several studies, Chinese adults (n = 466) showed the same tendency as American adults to honor implicit causal constraints when evaluating the plausibility of magical events. These findings suggest that graded notions of impossibility are shared across cultures, possibly because they are a byproduct of the interconnectedness of causal knowledge.

Similarity­-based theories of language processing predicts that the overlapping features (phonological,semantic, and pragmatic) of the to­be­integrated concepts in sentence could cause processing difficulties during online language understanding, particularly when the syntactic structure is complicated. Two noun phrases were embedded in object­relative vs. subject­relative clauses, and pragmatic similarity of the two phrases (similar: two male or two female proper names vs. dissimilar: one male and one female proper names) was further manipulated. Participants read the sentence in a self­paced moving window fashion and answered a comprehension question immediately after reading each sentence. Results revealed that neither similarity, syntax nor their interaction affected online measures of sentence processing, however, offline comprehension showed the predicted pattern that similarity impaired reading comprehension and this effect was exaggerated when syntactic structure was more complex. These results implied that similarity­-based theories of language processing might not apply to online integration of pragmatic information during complex sentence understanding. Further investigation with even more complex syntactic structures is under way to test this speculation.

Research on adults' numerical abilities suggests that number representations are spatially oriented. Dehaene et al. (1993) named such association of numbers with spatial response as the “SNARC (i.e., spatial–numerical association of response codes) effect,” wherein the right side (e.g. right hand) responds faster than left side when shown larger numbers (and vice-versa when shown smaller numbers).

Given that the SNARC effect serves as an automaticity index of cognitive process, we assume that the SNARC effect is influenced by familiarity. That is, compared to the numerical notation we are not familiar with (e.g. English words, Chinese characters), Arabic digits are more likely to trigger SNARC effect. To test this assumption, we introduced an adaptive procedure based on a simple perceptual orientation task that equates the mean reaction time difference between Arabic digits and traditional Chinese number words. Our results suggested that the SNARC effect interacted with notation, showing a SNARC effect for Arabic digits, but not for verbal numbers, challenging the view that notation does not affect numerical processes associated with spatial representations.

The SNARC effect can be generated in a relationship between space and other ordinal sequences––even weak ordinal information. We tested the SNARC effect in a color word sequence: Red, Orange, Yellow, Green, Blue, Indigo, and Violet (ROY-G-BIV) and found a reliable SNARC-like effect for Chinese color words, suggesting that without access to any quantitative information or exposure to any previous training, ordinal representation can still activate a sense of space. The results also support the inference that weak ordinal information without quantitative magnitude encoded in the long-term memory can activate spatial representation in a comparison task.