Conducting tourism research based on RSA: A case study on brand extension problem
Poster No:
833
Submission Type:
Abstract Submission
Authors:
Hang Yuan1, Siyang Luo1
Institutions:
1Department of Psychology, Sun Yat-sen University, Guangzhou, Guangdong
First Author:
Co-Author:
Introduction:
Tourism neuroscience primarily employs neuroscience technologies to explore individuals' behavior and neural activities in specific scenarios(Li et al., 2023). As an interdisciplinary field, it lacks integration of results from multimodal data. This research introduces representation similarity analysis (RSA) to studying tourism issues (Haxby et al., 2014). Using the example of brand extension, we demonstrate how to construct similarity models and conduct cross-modal comparisons.
Brand extension refers to a strategy wherein a brand with a certain level of recognition is applied to new products, aiming to reduce the market risks(Tauber, 1981). The evaluation of extension products determines whether it can achieve success in the market. Oxytocin, as a crucial hormone may enhance individuals' sensitivity to social cues(Groppe et al., 2013) and facilitate the recognition of social stimuli(Winslow & Insel, 2002). In this case, we conducted an experiment based on EEG and employed the RSA method to explore the potential mechanisms through which oxytocin influences consumers' attitudes toward brand extension products.
Brand extension refers to a strategy wherein a brand with a certain level of recognition is applied to new products, aiming to reduce the market risks(Tauber, 1981). The evaluation of extension products determines whether it can achieve success in the market. Oxytocin, as a crucial hormone may enhance individuals' sensitivity to social cues(Groppe et al., 2013) and facilitate the recognition of social stimuli(Winslow & Insel, 2002). In this case, we conducted an experiment based on EEG and employed the RSA method to explore the potential mechanisms through which oxytocin influences consumers' attitudes toward brand extension products.
Methods:
This study recruited 52 male participants (M = 20.13, SD = 2.296). A 2 (treatment: oxytocin(OT), placebo(PL)) × 2 (brand type: domestic, foreign) × 3 (perceived fitness: high, moderate, low) mixed-design was employed. During the experiment, participants were instructed to rate the fitness between brands and their extension products.
Based on preprocessed EEG data, representational dissimilarity matrixes (RDM) of different modalities were constructed, and the similarity between cross-modal matrices was compared. First, to examine the impact of oxytocin on neural activity patterns, the consistency of neural activity patterns within each group was calculated (Figure 1A). Second, to explore the influence of OT intervention on the relationship between neural activity patterns and conceptual models as well as behavioral patterns, behavior models were constructed based on participants' ratings of fitness. Theoretical models were built based on the similarity levels across stimuli. RDMs of neural activity were constructed over time points. Subsequently, the Mantel test (Mantel, 1967) was conducted at each time point to compare the similarity of RDMs.
Based on preprocessed EEG data, representational dissimilarity matrixes (RDM) of different modalities were constructed, and the similarity between cross-modal matrices was compared. First, to examine the impact of oxytocin on neural activity patterns, the consistency of neural activity patterns within each group was calculated (Figure 1A). Second, to explore the influence of OT intervention on the relationship between neural activity patterns and conceptual models as well as behavioral patterns, behavior models were constructed based on participants' ratings of fitness. Theoretical models were built based on the similarity levels across stimuli. RDMs of neural activity were constructed over time points. Subsequently, the Mantel test (Mantel, 1967) was conducted at each time point to compare the similarity of RDMs.
Results:
The t-test was employed to compare the differences in the neural activity similarities(r) between subjects in the OT and the PL group at each time point. The results revealed that within the 480~580 ms, participants in the OT group exhibited significantly higher consistency in neural activity patterns than PL group (Fig.1B).
Cross-modal comparisons between neural activity and behavioral RDMs for both the OT and PL groups indicated significant similarity between neural patterns and behavioral performance across multiple time windows (Fig.2B, D). It indicated that the relational patterns of neural activity were consistent with the patterns of behavioral performance. Furthermore, comparisons between neural activity RDM and conceptual RDM revealed that, in the OT group, within the time windows of 420~460ms and 590~650ms, participants exhibited significantly similar neural representation patterns in the occipito-temporal clusters to the conceptual model (Fig. 2C). In contrast, in the PL group, participants did not show significant similarity (Fig. 2E). These findings suggested that OT intervention enhances participants' neural representation of different types of stimuli.
Cross-modal comparisons between neural activity and behavioral RDMs for both the OT and PL groups indicated significant similarity between neural patterns and behavioral performance across multiple time windows (Fig.2B, D). It indicated that the relational patterns of neural activity were consistent with the patterns of behavioral performance. Furthermore, comparisons between neural activity RDM and conceptual RDM revealed that, in the OT group, within the time windows of 420~460ms and 590~650ms, participants exhibited significantly similar neural representation patterns in the occipito-temporal clusters to the conceptual model (Fig. 2C). In contrast, in the PL group, participants did not show significant similarity (Fig. 2E). These findings suggested that OT intervention enhances participants' neural representation of different types of stimuli.
·Figure 1. Gradient values and gradient differences between groups. (A) Primary-transmodal gradient and visual-sensorimotor gradient of rumination state at the IPCAS site. (B) Gradient differences in r
·Figure 2. Global gradient difference. (A) Gradient range difference in PKUSIEMENS site. (B) Gradient variance difference in PKUSIEMENS site.
Conclusions:
This study introduced the RSA analysis method into tourism neuroscience and conducted cross-modal analyses from a pattern perspective. Using the example of brand extension, we found that oxytocin not only enhances participants' neural representations of different extended brands but also promotes the consistency of their response patterns. In summary, conducting tourism studies using the RSA method can help us delve more profoundly into scientific questions in the field of tourism from a pattern perspective.
Emotion, Motivation and Social Neuroscience:
Social Neuroscience Other 1
Modeling and Analysis Methods:
EEG/MEG Modeling and Analysis 2
Other Methods
Physiology, Metabolism and Neurotransmission :
Pharmacology and Neurotransmission
Keywords:
Electroencephaolography (EEG)
Other - Representation Similarity Analysis (RSA)
1|2Indicates the priority used for review
Provide references using author date format
Haxby, J. V. (2014), 'Decoding neural representational spaces using multivariate pattern analysis', Annual Review of Neuroscience, 37, 435-456.
Li, S. (2023), 'A review of research into neuroscience in tourism: Launching the annals of tourism research curated collection on neuroscience in tourism', Annals of Tourism Research, 101, 103615.
Mantel, N. (1967). 'The detection of disease clustering and a generalized regression approach', Cancer Research, 27(2), 209-220.
Tauber, E. M. (1981). 'Brand franchise extension: New product benefits from existing Brand Names', Business Horizons, 24(2), 36-41.
Winslow, J. T. (2002), 'The social deficits of the oxytocin knockout mouse', Neuropeptides, 36(2-3), 221-229.