Dissertation research agenda
Trust and identification in AI-enabled selection
A multi-study, cross-cultural research program examining how applicants determine whether an organization—and the technology acting on its behalf—is worthy of trust and organizational identification.
Proposed theoretical model
From an experience to a relationship
Selection encounters generate justice judgments. Those judgments communicate relational value and organizational character, shaping trust and identification—and ultimately behavior.
→
→
→
Bibliometric foundation
The map of the field leads to the dissertation
The bibliographic-coupling study identified 410 documents and mapped the 100 most strongly connected papers. Its three research communities show where applicant-reactions scholarship began, how it explains psychological meaning, and where AI is pushing the field next.
Justice foundations
Selection fairness, job relatedness, opportunity to perform, explanations, treatment, and outcomes.
→
Psychological mechanisms
Attribution, uncertainty reduction, signaling, social identity, organizational identification, and trust.
→
AI and digital selection
Automation, opacity, privacy, innovativeness, reduced human contact, and applicant agency.
→
Trust × identification
Human, AI, and hybrid selection across time—and compared between Japan and the United States.
The missing connection
Justice evaluates the encounter. Trust interprets vulnerability. Organizational identification converts treatment into an identity-based relationship.
The mapping suggests that these mechanisms are adjacent but not yet sufficiently integrated—particularly when an AI system acts as the organization’s representative. This gap provides the field-level rationale for the proposed longitudinal, experimental, and cross-cultural studies.
Construct library
What each construct means—and how it enters the research
Each dossier separates the formal definition, core dimensions, dissertation role, and representative scholarship. Linked citations lead to the original paper or publisher record.
Applicant reactions
The applicant’s interpretation of the selection process
Working definition
Applicant reactions are the attitudes, perceptions, emotions, intentions, and behaviors that applicants develop before, during, and after the selection process. The construct shifts the unit of analysis from organizational actions to how applicants experience and interpret selection procedures.
Role in the proposed dissertation
The dissertation treats applicant reactions as the starting point in the selection process. Reactions should be measured at multiple stages—not reconstructed only after the selection decision—because interpretations of an assessment, interview, delay, or rejection can change over time.
Core dimensions and boundaries
- Perceived predictive validity and job relatedness
- Opportunity to perform and demonstrate relevant capability
- Consistency, transparency, explanations, and feedback
- Interpersonal treatment, two-way communication, and respect
- Affective reactions, test motivation, self-efficacy, and anxiety
- Organizational attractiveness, job-pursuit intention, offer acceptance, withdrawal, reapplication, and advocacy
Representative researchers and works
Established the organizational-justice model of selection-system fairness.
Developed and validated the Selection Procedural Justice Scale.
Synthesized applicant-reaction antecedents and outcomes through meta-analysis.
Organizational justice
Whether the process, outcome, and treatment are perceived as fair
Working definition
Organizational justice is the perceived fairness of organizational decisions and conduct. In selection, justice is not determined solely by whether an applicant is selected. It also concerns whether the procedure is job-related and consistent, whether the applicant receives dignified treatment, and whether the organization provides truthful and adequate explanations.
Role in the proposed dissertation
Organizational justice perceptions are modeled as the applicant’s evaluation of the selection process. They should predict organizational trust and organizational identification after controlling for outcome favorability—allowing the research to examine why some rejected applicants retain positive recommendation intentions while others do not.
Core dimensions and boundaries
- Procedural justice — fairness of the rules and decision process
- Distributive justice — fairness of the outcome and allocation
- Interpersonal justice — dignity, politeness, and respect
- Informational justice — candid, timely, and adequate explanations
Representative researchers and works
Helped establish organizational justice and distinguished the social side of fairness.
Validated the four-dimensional measure of organizational justice.
Translated justice principles into ten procedural rules for employee selection.
Group-value and relational models
Fair treatment communicates status, respect, and inclusion
Working definition
Group-value theory proposes that applicants value fair procedures partly because treatment by an organizational authority communicates whether they are respected as prospective members of a valued group. Relational models extend this insight: neutrality, trustworthy motives, and dignified treatment convey identity-relevant information about standing and inclusion.
Role in the proposed dissertation
This is the theoretical bridge between applicant reaction and organizational identification. A recruiter—or an AI system acting for the employer—does more than evaluate capability; it communicates whether the applicant is seen, valued, and potentially included.
Core dimensions and boundaries
- Standing — the status and respect communicated to the applicant
- Neutrality — unbiased, consistent, fact-based decision making
- Trustworthy motives — whether authorities appear benevolent and sincere
- Voice — a meaningful chance to present one’s perspective
- Pride and respect — identity-relevant judgments that connect treatment with cooperation
Representative researchers and works
Explained why procedures matter through relational judgments about authorities and groups.
Integrated procedural justice, identity, and cooperative behavior in the group-engagement model.
Organizational identification
A perceived sense of oneness with, or belonging to, the organization
Working definition
Organizational identification is the degree to which an applicant defines the self in terms of prospective membership in an organization. It is distinct from satisfaction, commitment, and organizational attractiveness: identification occurs when the organization’s successes, failures, values, and reputation become self-relevant.
Role in the proposed dissertation
The master’s thesis identifies organizational identification as the key psychological mechanism connecting applicant reactions with applicant outcomes. The dissertation will test when identification forms, whether trust precedes it, and whether rejection weakens—or respectful treatment preserves—it.
Core dimensions and boundaries
- Cognitive self-categorization — seeing oneself as an organizational member
- Oneness and belonging — perceived psychological connection
- Affective significance — emotional value attached to membership
- Evaluative significance — positive or negative value assigned to that identity
- Not interchangeable with organizational commitment, person–organization fit, or prestige
Representative researchers and works
Introduced social-identity theory to organizational research and clarified identification as self-definition.
Developed the widely used six-item organizational-identification scale.
Meta-analytically distinguished identification and examined its correlates and consequences.
Organizational trust
A willingness to accept vulnerability based on positive expectations
Working definition
Organizational trust is the applicant’s willingness to accept vulnerability to the organization based on positive expectations of its conduct. Trust is especially consequential in AI-enabled selection because applicants cannot directly inspect the data, algorithm, decision rules, or allocation of responsibility between human and machine.
Role in the proposed dissertation
Organizational trust may operate in parallel with organizational identification or serve as its antecedent: applicants may first assess whether the organization is worthy of reliance before incorporating prospective membership into the self-concept. The proposed studies will compare these models.
Core dimensions and boundaries
- Ability — competence to make a valid decision
- Benevolence — genuine concern for the applicant beyond self-interest
- Integrity — adherence to principles the applicant finds acceptable
- Predictability — consistency across applicants and over time
- Accountability — traceable responsibility, human oversight, review, and appeal
Representative researchers and works
Defined trust as willingness to accept vulnerability and specified ability, benevolence, and integrity.
Meta-analyzed trust in leadership, clarifying its antecedents and outcomes.
Connected trustworthiness directly to recruitment and selection research.
Signaling and AI-mediated selection
Every design choice reveals something about the employer
Working definition
Signaling theory explains how applicants form judgments when important organizational attributes cannot be observed directly. Applicants interpret recruiter conduct, assessment design, AI disclosure, response time, explanations, and appeal mechanisms as signals of organizational values and anticipated employee treatment.
Role in the proposed dissertation
The dissertation will distinguish the presence of AI from the way AI is designed and communicated. Human, AI-only, and hybrid processes will be compared under transparent versus opaque and respectful versus impersonal conditions.
Core dimensions and boundaries
- Signal content — innovation, competence, respect, or cost cutting
- Signal sender — recruiter, hiring manager, vendor, algorithm, or hybrid team
- Signal observability — whether AI involvement and criteria are disclosed
- Signal consistency — alignment among employer brand, policy, and actual treatment
- Signal credibility — verifiability, specificity, human accountability, and costly commitments
Representative researchers and works
Established signaling as a way to understand decisions under information asymmetry.
Reviewed validity and applicant reactions in digital-age personnel selection.
Examined AI-enabled recruitment through organizational attractiveness and applicant reactions.
Three-study program
Triangulating time, causality, and culture
No single method can establish the full argument. The program combines longitudinal applicant data, controlled experiments, and a matched Japan–U.S. comparison.
Longitudinal applicant study
Purpose: Observe how reactions, justice, trust, and identification change as real applicants move through the process.
- T1 — Application: prior familiarity, employer image, initial trust, and identification.
- T2 — Assessment/interview: procedural, interpersonal, and informational justice; AI awareness.
- T3 — Decision: outcome favorability, explanations, trust, identification, and intentions.
- T4 — Follow-up: acceptance, withdrawal, advocacy, word of mouth, and reapplication.
Larger multi-organization sample; temporal separation of measures; behavioral outcomes where available. Addresses small-sample, cross-sectional, and common-method limitations.
AI-selection vignette experiments
Purpose: Identify which process features cause trust and identification to rise or fall.
Human · AI · HybridTransparency
Explained · OpaqueTreatment
Respectful · ImpersonalOutcome
Accepted · Rejected
Randomized factorial designs will isolate AI use from poor implementation. Additional conditions can test human review, appeal, personalized feedback, and algorithmic-error correction.
Japan–United States comparison
Purpose: Test whether identical selection practices carry different relational meanings across institutional and cultural settings.
- Translate and back-translate matched measures and scenarios.
- Establish configural, metric, and scalar measurement invariance before comparing paths or means.
- Compare reactions to authority, AI autonomy, explanation, voice, and human accountability.
- Model labor-market and applicant-level differences rather than treating nationality as a complete explanation.
Culture is treated as a theoretically specified moderator—not a stereotype. The central comparison is whether the same signal produces different trust and identity judgments.
Proposed research questions
Questions that connect the studies
- How do applicant reactions and justice perceptions change across application, assessment, decision, and post-decision stages?
- Does organizational trust precede organizational identification, do they operate in parallel, or do they reinforce one another over time?
- Does organizational identification mediate applicant outcomes beyond trust, organizational attractiveness, and person–organization fit?
- How do human, AI-only, and hybrid decision systems change perceived ability, benevolence, integrity, and accountability?
- Can transparency, explanation, meaningful human oversight, appeal, and personalized feedback repair negative reactions to AI-enabled selection?
- Does rejection damage identification less when procedural, interpersonal, and informational justice remain high?
- Are the measurement structure and causal paths equivalent in Japan and the United States—and which cultural or institutional mechanisms explain differences?
- Which applicant reactions predict observable behavior: acceptance, withdrawal, recommendation, negative word of mouth, and reapplication?
Evidence boundaries
What the next research must improve
Combine perceptions with time stamps, recruiter records, decisions, acceptance, withdrawal, referrals, and later reapplication where feasible.
Recruit a larger, more heterogeneous sample across organizations, occupations, career levels, and selection technologies.
Measure antecedents, mediators, and outcomes at separate stages to test change and reduce retrospective reconstruction.
Pair field evidence with randomized vignettes that manipulate the selection agent, treatment, transparency, and outcome.
Demonstrate discriminant validity among identification, trust, attractiveness, commitment, and person–organization fit.
Establish measurement invariance and test explanatory moderators before interpreting national differences.
Selected references cited on this page
Foundational and methodological sources
- Aguinis, H., & Bradley, K. J. (2014). Best practice recommendations for designing and implementing experimental vignette methodology studies. Organizational Research Methods, 17(4), 351–371.
- Ashforth, B. E., & Mael, F. (1989). Social identity theory and the organization. Academy of Management Review, 14(1), 20–39.
- Bauer, T. N., Truxillo, D. M., Sanchez, R. J., Craig, J. M., Ferrara, P., & Campion, M. A. (2001). Applicant reactions to selection: Development of the Selection Procedural Justice Scale (SPJS). Personnel Psychology, 54(2), 387–419.
- Colquitt, J. A. (2001). On the dimensionality of organizational justice. Journal of Applied Psychology, 86(3), 386–400.
- Gilliland, S. W. (1993). The perceived fairness of selection systems. Academy of Management Review, 18(4), 694–734.
- Hausknecht, J. P., Day, D. V., & Thomas, S. C. (2004). Applicant reactions to selection procedures: An updated model and meta-analysis. Personnel Psychology, 57(3), 639–683.
- Mael, F., & Ashforth, B. E. (1992). Alumni and their alma mater. Journal of Organizational Behavior, 13(2), 103–123.
- Mayer, R. C., Davis, J. H., & Schoorman, F. D. (1995). An integrative model of organizational trust. Academy of Management Review, 20(3), 709–734.
- Tyler, T. R., & Blader, S. L. (2003). The group engagement model. Personality and Social Psychology Review, 7(4), 349–361.
- Vandenberg, R. J., & Lance, C. E. (2000). A review and synthesis of the measurement invariance literature. Organizational Research Methods, 3(1), 4–70.
- Woods, S. A., Ahmed, S., Nikolaou, I., Costa, A. C., & Anderson, N. R. (2020). Personnel selection in the digital age. European Journal of Work and Organizational Psychology, 29(1), 64–77.
This page presents a proposed dissertation framework. Relationships shown in the model are hypotheses to be tested, not established causal conclusions.