Poster #206 - Cross-modal interference in the primed categorization of emotion words and faces by children and adults.

• In Event: 1-005 - Poster Session 01
  In Poster Session: PS 01 Section - Social, Emotional, Personality

Integrative Statement

The multi-sensory perception of emotions from speech and facial expressions is a topic of much interest, as these two sensory modalities are almost always used simultaneously during in-person human social interaction.
Thus far, the multi-sensory nature of emotion perception from words and faces has been extensively studied in adults, but the developmental literature is still lacking in this area. Our experiments were thus aimed at exploring the degree to which information from emotional faces can influence the perception of emotion words, and vice versa, in 6-yeard old children (n=26, 14 female), 9-year-old children (n=25, 13 female), 12-year old children (n=23, 11 female), and adults (n=24, 12 female). We used a valence-based categorization task where participants were required to categorize the target stimuli (either words or faces) as being positive or negative as quickly as possible. Prior to the onset of the target stimulus, a prime from the opposite modality would be presented in congruent (prime and target are either both positive or both negative) and incongruent trials (prime is positive and target is negative, or vice versa). Experiment 1 involved the categorization of face targets preceded by emotion word primes, while Experiment 2 involved the categorization of emotion word targets preceded by face primes.
Our visual stimuli consisted of 48 photographs of emotional facial expressions made by 8 actors (4 female), with 24 of them being positive (including expressions of happiness and happy-surprise) and 24 negative (including expressions of sadness, anger, and fear). Our auditory stimuli consisted of 48 emotion words spoken by one male and one female trained speakers, including 24 positive words and 24 negative words. Both auditory and visual stimuli were balanced across the positive and negative categories in terms of their perceived arousal and valence.
For our analyses we used 4x2x2 (age x target-valence x congruency) ANOVAS to analyze both the accuracy and response times for correct trials in both experiments.
In Experiment 1 (word primes and face targets) we found that emotion word primes did indeed impact the categorization of face targets. The accuracy results showed a main effect of congruency (p < .01), with all participants making fewer errors in the congruent trials. An interaction between congruency and target-valence indicated that it was the incongruent trials with a positive target face which suffered from lower accuracy (Fig. 1). The analyses of response times for correct trials showed a triple interaction between age, target-valence, and congruency, indicating that unlike all other age groups, the 6-year-old children were faster in responding to trials with positive word primes regardless of congruency.
By contrast, the analyses of accuracy and response times from Experiment 2 (face primes and word targets) did not reveal any main effects or interactions of congruency.
Our results therefore demonstrate that emotion word primes influence the categorization of faces, particularly in younger children, but face primes do not influence the categorization of emotion words. The possible implications and mechanisms behind these findings will be discussed.

Group Authors

Collaborative Research Center - Cardinal mechanisms of perception: Prediction, Valuation, Categorization

Authors

• Michael Vesker, Justus-Liebig University Giessen
  Presenting Author

• Daniela Bahn, Philipps University of Marburg
  Non-Presenting Author

• Franziska Degé, Department of Developmental Psychology, Justus-Liebig-University Giessen, Germany
  Non-Presenting Author

• Christina Kauschke, Philipps University of Marburg
  Non-Presenting Author

• Gudrun Schwarzer, Department of Developmental Psychology, Justus-Liebig-University Giessen
  Non-Presenting Author