[page 10↓]

Summary

Previous research has revealed that the recognition of facial expressions and familarity may not be independent as postulated by traditional face recognition models. This dissertation attempts to localize this interaction within the information processing system by means of performance data and event-related potentials (ERPs). A simple paradigm was used in all experiments asking participants to perform a two-choice reaction time (RT) task either to discriminate facial expression or to discriminate facial familiarity. The respective task irrelevant dimension was varied independently of the task-relevant dimension (e.g. half of the presented faces in the expression discrimination task belonged to familiar faces).

Part I elucidates upon the question of whether there is an interaction between facial familiarity and the discrimination of facial expression. In Experiment 1, portraits of personally familiar and unfamiliar faces were categorized according to the emotional expressions happiness and disgust. Categorization was faster for portraits of personally familiar persons when compared to unfamiliar persons. This was especially pronounced for portraits displaying happiness. This advantage for familiar faces was not due to differential expressiveness of the portraits because it disappeared in participants for whom all portraits where unfamiliar. In Experiment 2 the same stimulus set was used as in the previous experiment. In addition, ERPs were recorded for 16 participants during the same expression categorization task. Although the weaker performance data of Experiment 1 were replicated, the peak latency of the N170 component of the ERP, reflecting structural encoding of the face, was not affected by familiarity. Also, the latency for the interval between the onset of the lateralized readiness potential (LRP) and the response (LRP-R), reflecting the duration of motor processes, was unaffected by familiarity. In contrast, the latency of the P300 component of the ERP, reflecting stimulus categorization time, and the interval between stimulus and LRP-onset (S-LRP), reflecting the duration of pre-motor processes, were shorter for happy familiar faces when compared to happy unfamiliar faces. Together the results suggest a facilitation of perceptual stimulus categotization for personally familiar faces displaying happiness. In order to elucidate upon the reduced effect in the RT of Experiment 2, another experiment was conducted with a slightly changed design. In addition, the skin conductance response was recorded to personally familiar and unfamiliar faces. This time, the facilitative effect of familiarity in performance data increased whereas it was not reflected by ERPs. Therefore, Experiment 4 and 5 used experimentally familiarized and unfamiliar faces in order to have a better control over the stimulus set. By discriminating happy from angry [page 11↓]faces (Experiment 4) or neutral from angry faces (Experiment 5) no facilitation was observed for experimentally familiarized faces. Hence, Experiment 6 used a set of stimuli consisting of famous and unfamiliar faces because semantic knowledge may be necessary for an interaction between facial familiarity and facial expression to emerge. Contrary to the hypothesis, no facilitation was observed for famous faces when discriminating neutral from happy faces. Together, the results of Part I imply a late perceptual but pre-motoric locus of the facilitative effect of familiarity on the discrimination of facial expression. Thus, the degree of familiarity may influence such an interaction since it was not observed for famous and experimentally familiarized faces. Different interpretations are discussed.

In Part II the question was raised whether there is also an interaction in the opposite direction. It was hypothesized that it would be possible to find an interaction between facial expression and the discrimination of familiar faces. Experiment 7 used the same personally familiar and unfamiliar faces as the first three experiments. Participants performed a familiarity discrimination task where they were shown portraits displaying happiness, disgust, or a neutral expression. Personally familiar faces were categorized faster as familiar if they displayed a happy or neutral expression. This advantage for happy and neutral familiar faces appears to be localized in the response selection stage as was suggested by an earlier onset of the S-LRP. In a final experiment participants performed a familiarity discrimination task on experimentally familiarized and unfamiliar faces. Again, it was hypothesized that facial expression has a facilitative effect on the discrimination of facial familiarity. However, no interaction was observed between facial expression and the discrimination of familiarity.

In summary, the results suggest that an interaction of facial familiarity and facial expression might be possible under some circumstances. Contrary to previous results this interinteraction is symmetrical because it was observed in the expression discrimination task as well as in the familiarity discrimination task. However, the degree of familiarity and the type of facial expression may be important for an interaction as it only emerged for personally familiar faces displaying happiness. Finally, the results are discussed in the context of possible interpretations, previous results, and face recognition models.


© Die inhaltliche Zusammenstellung und Aufmachung dieser Publikation sowie die elektronische Verarbeitung sind urheberrechtlich geschützt. Jede Verwertung, die nicht ausdrücklich vom Urheberrechtsgesetz zugelassen ist, bedarf der vorherigen Zustimmung. Das gilt insbesondere für die Vervielfältigung, die Bearbeitung und Einspeicherung und Verarbeitung in elektronische Systeme.
DiML DTD Version 3.0Zertifizierter Dokumentenserver
der Humboldt-Universität zu Berlin
HTML generated:
21.09.2004