Narrative/Constructivism for AI-based interviews: A Practical Application to Prepare Job Candidates

By Alan Jones and Nathan Mondragon

The changes to job interviewing resulting from COVID-19 have been transformative. During the lockdowns, employers shifted to video interviews to sustain their recruiting efforts. Many went further and adopted asynchronous (one-way) video platforms to find new, efficient methods of connecting with job candidates. Artificial Intelligence (AI)-based video interviews (AIVI) are a subset of this interview format. They use customized algorithms, suited for each job role, which

Interpersonal skills and role-specific traits are increasingly significant as a predictor of job performance and turnover. Companies with quantified business practices are allocating 30% to 40% of their hiring criterion to soft skills, perhaps reflecting the increasing formalization of these factors in their work environments (Bersin, 2018; Burning Glass, 2015). AIVI’s are designed to measure and report these competencies.

Additionally, decades of meta-analytic research on historical candidate screening methods such as degree obtained, schools attended, and GPA’s have shown to be poor indicators of performance once hired. Contemporary variables that are more predictive of on-the-job success (e.g., structured interviews and general ability tests) require larger scale data collection (Schmidt & Hunter, 1998). CV’s and resumes are no longer adequate for this task.

A job candidate’s answers to structured and behavior-based interview questions make up the input analyzed in this format. For individuals to perform well, maximizing and diversifying the language of their responses is a beneficial strategy (Butcher, 2021). AIVI’s do not simply compare a candidate’s word choice against a set of keywords and their proximal context like applicant tracking systems perform on resumes. Instead, they analyze a larger aggregate of spoken words and their intended meaning. The emphasis on language offers an opportunity to integrate the symbolism that narrative/constructivist-based practices generate. Micro-narratives, as defined by Savickas (2015), are an example. They are symbolic representations of concrete experience and provide possibilities for meaning-construction. They enable an individual job candidate to express their responses in AIVI’s in a clear and purposeful manner.

Safe Story-telling

A candidate’s personal stories are viable material for AIVI’s. Artificial intelligence, as an audience, does not judge life roles or personal situations. Candidates can safely elaborate on subjects that would be atypical by the social conventions of person-to-person interviews. The stories of family, cultural, political and religious affinities are acceptable content to use.

Normally, the candidate would be advised to avoid such content in their answer when a person is the interviewer. With an AIVI, the machine will not care. For example, if the words Republican or Democrat are used, then these terms could inject a clear bias into a human interviewer. It is important to note, most companies deploying AIVI’s still watch some of the video interview while simultaneously using the computer-generated scores to facilitate their screening decision. Over time, employers realize the AI scores are as good or better than their own reviews, so they pull back on reviewing all AIVI’s. It is a bit of an evolution in how employers become comfortable with the technology and rely more on it over time. Thus, the AIVI never fully replaces the human decision.

Candidates typically use the well-known STAR approach to respond to behavior-based questions. The formula is straightforward: state a Situation, Task or goal, followed by Actions taken, and the subsequent Results. With AIVI’s, candidates can elaborate beyond the traditional STAR response by describing their thought process and affective state-of-mind in the situation. Replacing the term, Task, with Obstacle, creating SOAR, highlights this purpose (See figure A.) This framework represents the candidate's early-stage uncertainty in problem-solving. More meaning-making can be gained in this section of a story by broadening one’s articulation here. An epilogue helps as well. By adding take-aways that demonstrate personal strengths, interests, and values, the candidate utilizes the allotted time and diversifies their inputted vocabulary (Ditewig-Morris, 2020). Career practitioners can share these tips with students to advance their candidacy.

Figure A - Job Candidates’ Response Framework

Jones Mondragon Figure A


Symbolism in Language

The average response time in the question/answer sequence of AIVI’s is three minutes. This provides enough time to produce a clear and concise answer while limiting the amount of unstructured rambling from less thought out responses. To make meaning succinctly, the candidate can employ the symbolism of the following common references.

These types of reflections generate specific language to use as qualifying phrases. They are also a good learning model for college students, as both Career Identity and Career Readiness are developed. In addition to familiarizing students with typical interview questions and advising them to develop their stories, career practitioners can help candidates enhance their performance with more meaning-making.

Non-Traditional Interviewing

AIVI’s reveal personal attributes that were previously elusive or hard to evaluate without injecting unconscious biases. Employers are successfully using them to select candidates with desired competencies and traits. This format is different from traditional interviews. It offers the candidate a chance to use one’s holistic self and can be an advantage. While STAR stories have been the traditional framework to craft interview responses, they are often rehearsed to some degree. A more open story-telling framework and the meaning-making devices recommended here lend themselves to revealing one’s personal intentions and to more imaginative use of language.



Amundson, N. E. (2010). Metaphor making. Ergon Communications.

Bersin, J. (2018, June 18). AI in HR: A real killer app. Forbes. https://www.forbes.com/sites/joshbersin/2018/06/18/ai-in-hr-a-real-killer-app/#2d26a86a48f1

Burning Glass. (2015.) The human factor: The hard time employers have finding soft skills. Burning Glass Technologies. https://www.burning-glass.com/wp-content/uploads/Human_Factor_Baseline_Skills_FINAL.pdf

Butcher, S. (2021). The new screening interviews used by investment banks. eFinancial Careers. https://www.efinancialcareers.com/news/2021/03/hirevue-interview-bank

Ditewig-Morris, K. (2020) STAR+E: An undergraduate teaching strategy to make the STAR model stick. National Association of Colleges and Employers. https://community.naceweb.org/blogs/kate-ditewig-morris1/2020/12/08/stare-an-undergraduate-teaching-strategy-to-make-t

Hartwell, M. & Chen, J. C. (2012). Archetypes in branding. How Books.

Leutner, F., Liff, J., Zuloaga, L., & Mondragon, N., (2021). HireVue’s assessment science. White Paper. https://webapi.hirevue.com/wp-content/uploads/2021/09/2021_08_24_HireVue_Assessment_Science_white_paper-Final-2.pdf?_ga=2.146560884.1756565331.1631578491-1540830016.1608752903

Savickas, M. L. (2015). Life design counseling manual. http://www.vocopher.com/LifeDesign/LifeDesign.pdf

Savickas, S. (2019). My career story universal: An autobiographical workbook for life-career success. http://vocopher.com/mcsu/MCS_26_GIUGNO%20(1).pdf

Schmidt, F. L., & Hunter, J. E. (1998). The validity and utility of selection methods in personnel psychology: Practical and theoretical implications of 85 years of research findings. Psychological Bulletin, 124(2), 262.


Alan JonesAlan Jones is Director, Career Center at Notre Dame of Maryland University. He can be reached at ajones17@ndm.edu.




Nathan MondragonNathan Mondragon is Chief I/O Psychologist at HireVue. He can be reached at nmondragon@hirevue.com

Printer-Friendly Version