Assessing the Formation of Experience-Based Gender Expectations in an Implicit Learning Scenario

Area of Science:

• Cognitive Science
• Psycholinguistics
• Artificial Language Learning

Background:

• Previous studies show explicit learning of gender cues in artificial languages leads to probabilistic expectations.
• These expectations might depend on overt awareness of gender cues during explicit learning.
• Investigating implicit learning is crucial to understand if awareness is necessary for expectation formation.

Purpose of the Study:

• To determine if experience-based expectations form without explicit awareness of a gender-coding system.
• To assess if implicit acquisition of gender cues leads to similar probabilistic expectations as explicit learning.
• To explore the timing of expectation emergence using eyetracking data.

Main Methods:

• An experiment using a gender-coded artificial language with spoken words and visual referents.
• Implicit learning scenario where participants acquired gender-coding without explicit instruction.
• Behavioral measures (recognition speed) and eyetracking data were collected.

Main Results:

• Participants developed frequency-based expectations comparable to explicit learning, despite imperfect gender-coding mastery.
• Referent recognition was faster when gender was consistent with induced expectations.
• Eyetracking data indicated expectations surfaced earlier in the implicit learning scenario.

Conclusions:

• Experience-based expectations are robust and not dependent on the manner of acquisition (explicit vs. implicit).
• Implicit learning effectively generates probabilistic expectations similar to explicit learning.
• Findings contribute to understanding the persistence of stereotype-like expectations in natural language processing.