Identification Of “faking” In Personality Tests: Evidence From Multimodal Dat

As we all know,the Big Five Model provides theoretical support for the development of personality questionnaire,and the importance of personality assessment has gradually manifested in lots of research fields.Considering the strong predictive effect of personality on employees’job performance and other benefits brought by person-job fit,many enterprises include personality assessment as one of the important criteria for personnel selection.Ideally,the personality assessment can help a lot in personnel selection,but the actual situation is not optimistic.Out of the pursuit of high-risk results,many applicants faking in the assessment and showing favorable self-representation deliberately to obtain the job,it will not only lead to the deviation of personality assessment,but also destroy the fairness and accuracy of personnel selection decisions.In order to avoid the serious consequences of faking,there is an urgent need for effective methods to identify it.Traditional methods often use scores and response latency to identify faking,but they are easily distorted.Physiological parameters,brain activation and other measurement indicators that are not easily distorted can provide necessary guarantee for the identification of faking.Although those indicators have their own values,the scores are prone to be artificially distorted,the response latency,physiological parameters and brain activation need additional professional technical support.In contrast,person-fit statistics（PFS）,which are used to identify aberrant response patterns,have the advantages of low cost,high efficiency and no equipment restriction.Therefore,the current study aimed to find out the best PFS that can effectively identify faking,which supported by the multimodal data of score,response latency,physiological parameters and brain activation.This study rooted in the experimental faking-induced paradigm,polygraph and f NIRS were used to record physiological parameters and brain activation when subjects complete the Big Five Inventory-2 and the Faking Detection Scale with mouse,,₉),3and7)were calculated to explore the best PFS that can identify faking effectively.The current study found that（1）People can fake in all the dimensions of BFI-2,conscientiousness and neuroticism dimensions were important in faking identification.（2）Response latency under faking condition was shorter than honest condition,which supported the semantic processing model.（3）Faking would lead to a significant increase in individual heart rate and finger pulse rate,as well as the brain activation in prefrontal cortex and inferior parietal lobule.（4）The identification effect of,₉),3and7)were not ideal.（5）There was a close relationship between the scores of Faking Detection Scale and faking behavior.Conclusion:The identification effect of,₉),3and7)were not ideal,but Faking Detection Scale can still provide good guarantee for faking identification.