Individual Differences in Distributional Learning for Speech: What's Ideal for Ideal Observers?

Area of Science:

• Psycholinguistics
• Auditory Perception
• Cognitive Science

Background:

• Speech perception relies on dynamically adjusting sound mappings to accommodate variations in input.
• Categorical perception of speech is influenced by the variability of speech sound distributions.
• Understanding the link between higher-level language abilities and adaptation to low-level speech cues is crucial.

Purpose of the Study:

• To investigate the hypothesis that receptive language ability is associated with the capacity to adapt to distributional cues in speech.
• To examine how individual differences in receptive language affect distributional learning in speech perception.

Main Methods:

• Fifty-eight listeners participated in a distributional learning task involving phonetic categorization of /g/ and /k/ sounds.
• Two blocks of the task used voice onset time distributions with narrow (consistent) and wide (variable) variances.
• Receptive language ability, nonverbal intelligence, and reading fluency were assessed.

Main Results:

• Participants demonstrated distributional learning, showing more categorical perception with consistent input compared to variable input.
• Individual variability in the magnitude of distributional learning was observed.
• Receptive language ability, but not nonverbal intelligence or reading fluency, predicted the extent of this learning.

Conclusions:

• Individual differences in speech distributional learning are significantly linked to receptive language ability.
• Weaker receptive language skills correlate with a reduced ability to utilize consistent input distributions for speech perception.
• This suggests a role for higher-level language processing in adapting to low-level acoustic variations in speech.