Date of Award

Summer 8-2026

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Marketing

Program/Concentration

Business Administration - Marketing

Committee Director

Mahesh Gopinath

Committee Member

Yuping Liu-Thompkins

Committee Member

Andrew Bennett

Abstract

Across two essays, this dissertation demonstrates that mimicry and personality shape consumers’ perceptions of chatbot intelligence, creativity, and fit, with downstream effects on recommendation acceptance that depend on factors such as mimicry intensity, brand familiarity.

Essay1: Mimicry, a central aspect of human interaction, is increasingly being examined in human–AI communication, particularly with chatbots. While prior research shows mimicry can build rapport, trust, and engagement, it has often been treated as a binary factor, overlooking how different levels may influence user perceptions. This research proposes and tests a dual-mediation framework in which different levels of chatbot mimicry affects recommendation acceptance through perceived intelligence and perceived creativity. Across four experiments, this study examines how mimicry level (low, medium, high), mimicry type (informational vs. emotional), and prior chatbot experience shape these outcomes. Study 1 tested the main effect: high mimicry reduced perceived intelligence, perceived creativity, and recommendation acceptance relative to medium mimicry. High levels of mimicry reduced chatbot recommendation acceptance, while low and medium levels of mimicry did not differ from each other. Study 2 identified perceived ease of use as a boundary condition, showing that the negative effects of high mimicry were particularly pronounced among users who found chatbot interactions easy and intuitive. Study 3 examined whether mimicry type moderates these effects. Informational mimicry produced higher perceived intelligence than emotional mimicry at medium levels but not at high levels. For perceived creativity, high emotional mimicry led to higher ratings than medium emotional mimicry. For recommendation acceptance, medium informational mimicry outperformed medium emotional mimicry, suggesting that reflecting factual content is more beneficial than emotional content at moderate mimicry levels. Study 4 replicated the main effect of mimicry level and tested prior chatbot experience as a moderator. Prior experience did not moderate the indirect effects through perceived intelligence or perceived creativity, providing no support for the hypothesized moderation. These findings demonstrate that chatbot mimicry effects are nonlinear, context-dependent, and sensitive to mimicry type, offering practical guidance for chatbot designers to calibrate mimicry levels and content accordingly.

Essay 2: As chatbots increasingly act as brand representatives, the personality they express may need to fit the personality of the brand. This research tests how the fit between chatbot personality (introverted vs. extraverted) and brand personality (sincere vs. sophisticated) shapes perceived fit and recommendation acceptance across three studies. Study 1 used fictitious brands, Study 2 used real and familiar brands, and Study 3 used two sophisticated real brands to test brand awe as an explanation for the brand effect found in Study 2. Perceived fit was higher when the chatbot matched the brand, but only for fictitious brands, and this congruence did not carry over to recommendation acceptance in any study. Instead, the introverted chatbot was preferred for fictitious brands and the extraverted chatbot for familiar ones, which suggests that brand familiarity may decide which chatbot style works best. In addition, brand awe raised both perceived fit and recommendation acceptance. Perceived fit and recommendation acceptance did not move together, which sheds light on a literature that has assumed fit leads to better outcomes and shows that the two have partly different drivers. The theoretical and managerial implications of these findings are discussed.

Rights

In Copyright. URI: http://rightsstatements.org/vocab/InC/1.0/ This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s).

DOI

10.25777/jec6-eg81

ISBN

9798193214212

ORCID

0000-0002-2198-1202

Download

Included in

Marketing Commons

Share

COinS