Thinking new thoughts: How problems constrain the search for their solutions
Problems pervade us. Yet in science and everyday thought, we solve countless problems, learning and discovering something new along the day. While much work has looked at how evidence affects theory change and belief revision, there are many problems for which evidence is as yet unobserved or perhaps unobservable. It remains an open question how we identify, generate, and evaluate new ideas prior to receiving evidence. We suggest that the problem itself may give clues about what is desirable in the solution -- clues which impact the generation and evaluation of hypotheses. In this talk we present a series of experiments investigating how adults and children evaluate conjectures for ad hoc problems. We end by proposing some future directions for thinking about problems.
The language network is recruited but not required for non-verbal semantic processing
Consistent with longstanding findings from neuropsychology, several brain regions in left frontal and temporal cortex respond robustly and selectively to linguistic information. But are these regions selectively engaged in processing linguistic meaning? Or do they instead store and process semantic information independent of its format (pictures versus words)? To address this question, in Experiment 1 we scanned participants with fMRI while they performed a semantic task vs. a perceptual control task on sentences and line drawings that describe or depict simple agent-patient interactions. The language regions showed above-baseline response to semantic processing of both sentences and pictures, demonstrating their engagement during non-verbal semantic processing. But is this engagement necessary for understanding pictorial stimuli? In Experiment 2, we tested two individuals with global aphasia on the semantic task on the same picture materials, as well as a sentence-picture matching task. As expected, individuals with aphasia were at chance in the latter. However, they did not differ from the controls in performance on the semantic judgment task, indicating that the left fronto-temporal language system is not necessary for processing non-verbal meanings.
Audiovisual Synchrony Preference and Prefrontal Cortical Activity in Infants
Young infants are sensitive to the statistical relationship between visual and auditory input from their environment. For example, when watching videos of a person talking, infants prefer to look at a synchronized video versus one where the auditory input does not match up with the visual input (Dodd, 1979; Kuhl & Meltzoff, 1982; Patterson & Werker, 1999). Individual differences in synchrony preferences are reliable and predict other measures of infants’ cognitive development (Bahrick et al, 2018). However, little is known about the cognitive or neural mechanisms underlying this preference. We hypothesize that this synchrony preference reflects individual differences in detecting and encoding statistical structure in the input. To test this hypothesis, we will ask whether infants with stronger preferences for synchrony also show better learning of statistical structure within a single modality and greater activation of lateral prefrontal cortex during the initial encoding phase. We use non-invasive functional near-infrared spectroscopy to record changes in blood oxygenation in lateral and medial prefrontal cortex in 8-10 month old infants. At cog lunch, I will present the experimental design for this study as well as pilot data attesting to the feasibility of the design.