Factors affecting the utility of emotional stimuli in research

Diconne, Kathrin (2023) Factors affecting the utility of emotional stimuli in research. Doctoral thesis (PhD), Manchester Metropolitan University.

Preview

Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (1MB) | Preview

Abstract

Emotional stimuli such as images, words, or video clips are often used in studies researching emotion. These stimuli are provided to the research community in sets accompanied by normative rating data indicating the emotional value of each stimulus. With emotional stimuli sets continuously being published, an immense number of available sets are complicating the task for researchers looking for suitable stimuli. Therefore, a systematic review was conducted to find all existing emotional stimuli sets that are freely available or available upon request. The result was the creation of the KAPODI-database containing 364 sets and presenting a comprehensive list of set characteristics. A searchable online version allows researchers to find and compare individual sets, as well as to add new published sets. Previous research has shown that factors such as assessors’ age, gender, or ethnicity, influence stimulus perception. Nevertheless, researchers often rely on the provided normative rating data without verifying its validity for their own participant sample. Additionally, findings regarding the effect of emotions onto memory are inconsistent, with sometimes enhancing, and sometimes detrimental effects. A possible reason for these contradictory results could be factors influencing stimulus validity that have yet not been investigated. Therefore, two additional studies were conducted. A first study sought to analyse these possible factors by investigating validity of stimuli in relation to assessed dimension, namely valence (negative to positive), arousal (calming to exciting), and dominance (no dominance to high dominance), as well as different dimension categories (e.g., low/medium/high valence, arousal, and dominance, respectively), and standard deviation (SD) categories (low/medium/high) both for images and words. In the second study, the factor of sensory processing sensitivity (SPS) that is known to positively correlate with the depth of processing of emotional content was investigated. In this latter experiment perception as well as episodic recognition of emotional image stimuli were assessed and analysed in relation to level of SPS. The two experimental studies suggest that solely valence seem reliable for both image and word stimuli, while arousal, dominance, dimension, and SD category are not reliable. Moreover, perception of emotional stimuli differs between individuals of low vs. high SPS regarding low-valence stimuli only, with high SPS individuals perceiving these stimuli as more negative. Finally, recognition of stimuli increased with increasing arousal, and decreased with increasing valence. Together, these results urge researchers to validate arousal and dominance ratings of selected stimuli for their participant sample prior to study conduction, as well as to consider the participants’ sensitivity if the study uses negative (low-valenced) stimuli.