Skip to main content

Predicting Suicide

Periodic Reporting for period 1 - PS (Predicting Suicide)

Reporting period: 2018-07-30 to 2020-07-29

Suicide is a leading cause of death worldwide. According to the World Health Organization, approximately 800,000 people die by suicide each year, and estimates of 20 times this number make suicide attempts. Prejudice/stereotyping can also be extremely damaging to large groups of individuals. Importantly, these behaviors have a profound societal cost and can severely impact families and communities. However, to change these behaviors, it is crucial that we precisely know the mental associations (internal psychological factors) and environmental factors (external) that can exacerbate or ameliorate these behaviors. One of the primary goals of this fellowship was to determine whether a weakened association between “Me” and “Life”, a stronger association between “Me” and “Death”, or both these associations are crucial for differentiating suicide attempters from non-attempters.

We showed that a weakened association between “Me = Life” is more strongly predictive of having a history of suicidal attempts than is a stronger association between “Me = Death/Suicide.” However, among those who previously attempted suicide, a strengthened association between “Me = Death” is more strongly predictive of the recency (and frequency) of a suicide attempt. These results suggest that decomposing traditional IAT D-scores can offer new insights into the mental associations that may underlie clinical phenomena and may help to improve the prediction, and ultimately the prevention, of these clinical outcomes. We are now applying this new technique to an in-patient clinical sample (N > 2,000) and to an adolescent sample gathered at ER wards (N > 6,000) to determine if it can prospectively predict completed suicide/attempts.

Racial tensions in the U.S. have recently increased, and understanding both individual and environmental factors that relate to negative intergroup relations are important for implementing systemic societal changes. We predicted that people living in regions (U.S. states) with higher infectious disease rates have a greater tendency to avoid out-groups because such avoidance reduces their perceived likelihood of contracting illnesses. Consistent with this parasite-stress hypothesis, we show that both White individuals (N > 770,000) and Black individuals (N > 150,000) living in U.S. states in which disease rates are higher display increased implicit (automatic) and explicit (conscious) racial prejudice. These findings remained robust even after accounting for a whole host of individual and environmental factors. Furthermore, white individuals with high germ aversion tendencies were especially prejudiced when reminded of viruses/diseases/infections. These findings could also account for the rise in discrimination, targeting minority groups in the U.S. (e.g. Asian or Black Americans) since the COVID-19 outbreak. This research may also generalize to other countries where intergroup tensions are apparent.
When I started the project, I spent a large portion of time learning how to code online reaction time experiments. I have learned how to use R, and I have since developed multiple scripts to clean and analyze the data on PI and PIH. I also learned how to use cloud computing in order to clean the 2019 Race IAT task for PI, so it could be posted for public use on OSF. I am often asked for assistance with cleaning and scoring data from implicit measures, and this has resulted in many fruitful collaborations with various clinical psychologists and psychiatrists. I also learned how to create my own website to promote my academic profile (https://psychologyboss.com/ still under construction) and how to perform multinomial modeling techniques, such as Quad modeling.

This fellowship has so far supported three first-author journal articles published or accepted for publication in internationally recognized (tier 1) journals, and another second author journal article. I have two other first author publications that are under review and at three others that I am affiliated with currently under review at high impact journals (e.g. Journal of Experimental Psychology: General and the American Journal of Psychiatry). I expect to have two to three more first-author publications under submission by the end of the fellowship related to the impact of COVID-19 in the social and clinical implicit cognition domains.

Apart from traditional journal publications, I have also been quite successful at disseminating my Marie Curie research. I chaired and was the sole organizer of a symposium at the largest and most competitive psychology conference in 2019 (Association for Psychological Science). The title of the symposium was: “Advances in Implicit Cognition: New Measures, Model Investigations, and Statistical Techniques”. Speakers included a colleague at Harvard University (Tessa Charlesworth), Franziska Meissner (University of Jena), and Adriaan Spruyt (Ghent University), founder of https://implicitmeasures.com/. Follow up meetings with Dr. Meissner allowed me to make contact with Prof. Klaus Rothermund. We currently have a German Humboldt fellowship submitted together along with Prof Christoph Klauer (University of Freiburg) that aims to begin after my current fellowship.

I have also presented talks on my research at a number of conferences and impactful labs (e.g. SPSP, Bethany Teachman’s PACT Lab). My research has also been picked up by various international media outlets (e.g. NPR, Harvard Gazette, Pacific Standard, Psychology Today ). I was also invited to present my research at Harvard Medical School, which included discussion on the Marie Curie fellowship and how scholars can acquire it. Moreover, I am a PIH executive board member, and I am also the liaison officer between PIH and PI, so I have some influence regarding the direction of the world's largest online psychological laboratory. Lastly, in March 2020, I was appointed to the role of the Director of the PI International sites, so this position has hugely expanded my collaborative reach.
I have covered many of these points in the previous two questions. However, I will run one more experiment validating the SIP in the political domain (Biden versus Trump) and compare the results to the IAT. Following this study, I will submit a manuscript to the Journal of Personality and Social Psychology. I will also run a study with the SIP to test why an intervention (inference based cognitive behavioral modification) aimed to increase alcohol abstinence during “Dry January” was effective or ineffective for certain participants. I will supervise M.Sc. students at the University of Amsterdam on projects analyzing the big data made available to me by PI.

To change and improve society, we must first understand the primary factors that are leading to negative outcomes. The new methodological techniques I developed are crucial to enhancing our understanding of biases towards separate attitudes objects (e.g. Me-Life versus Me-Death). Moreover, by using big data, I have been successful at demonstrating the crucial “environmental” (i.e. infectious disease) and individual level (i.e. germ aversion) factors that can partly account for racial tensions in the U.S. All these findings have the strong potential to assist with reducing suicide rates and prejudice in society. In conclusion, my Marie Curie Global Fellowship was instrumental in advancing knowledge in both clinical and social psychological domains.
eu-photo.png