We use our senses to navigate in the world. An important property of olfaction, the sense of smell, is to enable us to approach beneficial things, and to avoid what might be toxic or otherwise harmful in our environment. Other peoples’ behaviors are also paramount for our survival. Thus, we use our vision to decode their internal states from their facial expressions. For many modalities, multiple senses are integrated to enhance sensory percepts. In this thesis, I investigated how valenced odors affect the perception of facial expressions. Specifically, using a multi-method approach, I studied the integration of unpleasant and pleasant odor contexts on odor-congruent and incongruent facial expressions, disgusted, and happy faces.
The effects I am interested in are those that valenced odors have on face perception, attention to faces, and the cortical processing of faces. To answer these questions, I used questionnaires, ratings, EEG, and behavioral measures such as reaction times. Across studies, ratings of face valence are affected in the direction of the odor valence (e.g., faces are rated more negatively in the context of an unpleasant odor). Also, overall, the results in my studies indicate that faces are perceived as more arousing in valenced odor contexts; however, these effects occur regardless of facial expression.
In study 1, I found that valenced odors and facial expressions are integrated at an earlier time-frame than previously thought. Specifically, I found that the N170 event-related potential component (ERP) to disgusted facial expressions was lower in amplitude in the unpleasant odor condition than in the pleasant odor condition. This effect was not present for happy faces in the N170 component. An unpleasant odor might thus facilitate the processing of threat-related information.
In study 2, I found evidence that odors, in general, did not affect the recognition speed of facial expressions that changed from neutral to disgusted or happy over 3 seconds. Also, I found robust evidence against congruency effects in facial expression recognition reaction times (RTs). The results indicated that faces overall were recognized faster in the unpleasant odor condition. Further, these results were not qualified by individual differences in body odor disgust. Thus, unpleasant odors might facilitate the recognition of facial expressions regardless of trait body odor disgust.
In study 3, I studied whether valenced odors directed spatial attention toward odor-congruent facial expressions in a “dot-probe” task. I found decisive evidence that odors do not affect attention towards disgusted and happy facial expressions, casting doubt on the dot-probe experiment. However, I found that probes were detected faster as a function of time-on-task in the unpleasant odor condition. I hypothesized that this effect might be due to maintained vigilance in the presence of an unpleasant odor and task fluency effects.
In summary, the results indicate that valenced odors affect facial perception. Generally, faces are perceived as more valenced and arousing in odor contexts. Further, an unpleasant odor may decrease RTs; however, this effect seems to be irrespective of the target type. Also, odor face integration may happen earlier than thought; yet, evidence in the literature is mixed, and more research is needed. The methods I have used may increase transparency and robustness of published results, and help accelerate knowledge development in this field of research.