baueraj / event-segment-video_behaviour

Analysis of behavioural response data showing how adults vs. children segment their continuous experience of the world into discrete events

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

event-segment-video_behaviour

Analysis of behavioural response data showing how adults vs. children segment their continuous experience of the world into discrete events. Specifically, participants segmented their experience of watching videos into separate events.

Code description

Data preparation/cleaning and analysis showing how children vs. adults segment their continuous experience of watching cartoons into discrete events -- the first step to encoding experience into distinct episodes in long-term episodic memory. Response data include timestamps of when participants identify an event boundary in a cartoon, and their explanations for why the timestamps correspond to times when one event ends and another begins.

Research description

Neural development of the perception and memory of event structure in continuous narrative

The brain is remarkable for its ability to extract percepts out of a deluge of sensory input, but perhaps more impressive is its capacity for segmenting continuous experience into meaningful units that are remembered and define the story of one’s life. Everyday life consists in knowing how to navigate a variety of events, such as ordering food at a restaurant, which requires breaking an event down into its components and understanding how they meaningfully relate to each other. Previous research suggests that adults segment events in a way that incorporates perceptual details with knowledge of similar past experiences. However, it remains unclear how well children are able to integrate these sources of information in carving up their experiences. For example, given their impoverished knowledge of the world, do children often fail to recognize shifts in an agent’s goals, thereby perceiving and encoding events that are less structured? The current project examines how children (compared to adults) perceive and remember event structure in continuous video narrative, a medium that mirrors everyday experience. Furthermore, this research will apply machine learning techniques (including hidden markov modeling) to fMRI data to reveal how the developing brain segments and encodes events.

About

Analysis of behavioural response data showing how adults vs. children segment their continuous experience of the world into discrete events


Languages

Language:Python 100.0%