The Clinical Angle - Dombrovski
What does the research on decision-making in suicide tell us about prevention opportunities?
Alexandre Y. Dombrovski
University of Pittsburgh Department of Psychiatry
Clinicians have long observed that the decision to commit suicide is often made after a limited and shallow consideration of the current crisis, possible consequences, and alternative solutions. Accumulating empirical evidence indicates that people who engage in suicidal behavior are impaired in their ability to judge the consequences of their choices. This notion is suggested by the overlap between suicidal behavior on one hand and addiction and gambling, on the other. Early laboratory studies of suicide attempters uncovered their poor performance on gambling tasks (Clark et al., 2011; Jollant et al., 2005). What might explain it? The expected consequences or, formally speaking, expected value of one’s choices are represented in the ventromedial prefrontal cortex, and one hypothesis is that such representations are disrupted in people who attempt suicide. Indeed, functional brain imaging augmented with a computational model revealed that impaired decision-making in suicide attempters is paralleled by disrupted expected value signals in the ventromedial prefrontal cortex (Dombrovski et al., 2013). Behavioral studies have linked an increased preference for immediate rewards (delay discounting) to low-lethality/poorly planned attempts (Dombrovski et al., 2011), multiple attempts (Dougherty et al., 2004), and the co-occurrence of attempted suicide and addiction (Liu et al., 2012). On the other hand, well-planned, serious suicide attempts were associated with intact or even diminished delay discounting (Dombrovski et al., 2011). The latter observation points to the existence of distinct pathways to suicide, marked by different decision deficits and probably also by different stressors.
Of particular interest to the readers of this newsletter may be how people prone to suicide make social decisions. As we know, precipitants and deterrents to suicidal acts are often of a social nature, such as conflict in the former case and the impact on family, in the latter. Whereas existing frameworks such as the interpersonal theory focus on precipitants, the studies discussed above suggest that people who engage in suicidal behavior are also impaired in their ability to represent the consequences of their decisions. One of the tools for examining the mechanisms and individual differences in social decision-making is the study of social exchanges of the type described by game theory. For example, Szanto and colleagues examined the behavior of older depressed individuals on the ultimatum game, where a “proposer” divides a pie between herself and a counterpart, “the responder”, the role played by the subject in the study, who in turn can accept the split or reject the offer. When the offer is rejected, neither gets anything. People often reject very unfair offers in one-shot exchanges, even though this behavior confers no tangible advantage, but they are less likely to reject offers with a higher stake (e.g. $50 vs. $5), showing sensitivity to the consequences of their choice. This sensitivity was abolished in high-lethality suicide attempters, suggesting an inability to weigh the consequences (in this case, for themselves) of their social decisions (Szanto et al., 2014).
One practical implication of this research for prevention is that, if indeed suicide is attempted without proper consideration of the consequences, the window of risk will often be limited to the current crisis. In our longitudinal study of attempted suicide in late-life depression three-quarters of the participants say that they regret their suicide attempts. This view resonates with the conceptualization of suicide as an accident by Jan Beskow, who studied suicides on the Swedish railroads and found that determinants of survival were quite similar between suicides and accidents (Beskow et al., 1994). Just like the accident occurs with some probability when demands (e.g. road conditions) exceed performance (e.g. driving skills), suicide may take place with some probability when the impact of stressors and availability of means prevail over coping skills and deterrents. Small increases in performance or decreases in demands may thus be effective in preventing suicide, if achieved at the right time. This approach to suicide prevention is reinforced by the success of means restriction, such as limiting access to firearms or toxins. It is similarly not surprising that bridging interventions for high-risk patients in the emergency room are perhaps the only treatment strategy shown to prevent death by suicide (Riblet et al., 2017). These interventions typically involve an interview with the patient in an emergency care setting followed by brief contact with the aim of helping one connect with outpatient treatment. One may be astounded that such a limited intervention can have an impact on the decision of whether or not to take one’s life. Yet, most patients experience a deep ambivalence about suicide, and a slight nudge at the right time may cause the deterrents and alternative solutions to prevail.
A more difficult question is what specifically can be done to prevent an interpersonal crisis from escalating to suicide in a person prone to making bad decisions. Existing psychotherapies, notably learning-based therapies and mentalization-based therapy, have been shown to address this problem to some extent (Riblet et al., 2017), building on the insights of learning and psychodynamic theories of the mid-20th century. Yet, their real-world effectiveness is hampered by limited availability, and the necessary commitment is often resisted by the most vulnerable patients. Web-based therapy may prove to be an effective alternative medium, but it has yet to overcome the problem of commitment. These challenges highlight the need for new brief, targeted interventions. Fresh ideas are needed here. One new theoretical insight from decision neuroscience that may inform new approaches is that maladaptive responses to social stressors may stem from Pavlovian (stimulus-outcome, S-O) motivational influences on goal-oriented (response-outcome, R-O) behavior. This relates to two features of socially motivated self-destructive behavior: its insensitivity to adverse consequences and heightened sensitivity to internal emotional state (King-Casas et al., 2008; Sadikaj et al., 2013; Szanto et al., 2014). These are key properties of the Pavlovian (S-O) system, distinguishing it from the R-O systems involved in goal-directed action (Balleine and Dickinson, 1998; Domjan, 2005). Hence, social stressors may act as potent Pavlovian conditioned cues that interfere with adaptive learning and decision-making. This interference emerges when the emotional context pushes the patient toward an action with potentially aversive consequences, suicide being an extreme example. If, for example, one is angry with a seemingly distant partner, the Pavlovian response congruent with the angry internal state is to start an argument, leading the partner to withdraw. As this scenario is repeated, one could learn that fighting pushes the partner away (response-outcome) and restrain oneself. Yet, people with vulnerabilities such as borderline personality disorder, may learn to associate the partner or his/her communications (the conditioned stimulus) with the outcome of abandonment. As fights continue and the partner becomes more aloof in response [cf.(Sadikaj et al., 2013)], the crisis can culminate in a suicidal act. In other words, aggressive, hostile and self-destructive reactions to social conflict may constitute Pavlovian conditioned responses. One direction suggested by this theory is emphasizing the consequences of suicidal behavior, something experienced therapists do with suicidal patients. Could a public education campaign or even social advertising be similarly effective?
Michael Hallquist (Penn State Univ.) helped me develop ideas about Pavlovian influences on goal-oriented behavior. For a more detailed discussion of decision biases in suicide, please see (Dombrovski and Hallquist, 2017).
Balleine, B.W., Dickinson, A. (1998). Goal-directed instrumental action: contingency and incentive learning and their cortical substrates. Neuropharmacology, 37, 407–419.
Beskow, J., Thorson, J., Öström, M. (1994). National suicide prevention programme and railway suicide. Social Science and Medicine, 38, 447–451.
Clark, L., Dombrovski, A.Y., Siegle, G.J., Butters, M.A., Shollenberger, C.L., Sahakian, B.J., Szanto, K. (2011). Impairment in risk-sensitive decision-making in older suicide attempters with depression. Psychology of Aging, 26, 321–330.
Dombrovski, A.Y., Hallquist, M.N. (2017). The decision neuroscience perspective on suicidal behavior: evidence and hypotheses. Current Opinion in Psychiatry, 30, 7-14.
Dombrovski, A. Y., Szanto, K., Clark, L., Reynolds, C. F., & Siegle, G. J. (2013). Reward signals, attempted suicide, and impulsivity in late-life depression. JAMA psychiatry, 70, 1020-1030.
Dombrovski, A. Y., Szanto, K., Siegle, G. J., Wallace, M. L., Forman, S. D., Sahakian, B., ... & Clark, L. (2011). Lethal forethought: delayed reward discounting differentiates high-and low-lethality suicide attempts in old age. Biological psychiatry, 70, 138-144.
Domjan, M. (2005). Pavlovian conditioning: a functional perspective. Annual Review of Psychology, 56, 179-206.
Dougherty, D. M., Mathias, C. W., Marsh, D. M., Papageorgiou, T. D., Swann, A. C., & Moeller, F. G. (2004). Laboratory measured behavioral impulsivity relates to suicide attempt history. Suicide and Life-Threatening Behavior, 34, 374-385.
Jollant, F., Bellivier, F., Leboyer, M., Astruc, B., Torres, S., Verdier, R., ... & Courtet, P. (2005). Impaired decision making in suicide attempters. American Journal of Psychiatry, 162, 304-310.
King-Casas, B., Sharp, C., Lomax-Bream, L., Lohrenz, T., Fonagy, P., & Montague, P. R. (2008). The rupture and repair of cooperation in borderline personality disorder. science, 321, 806-810.
Liu, R. T., Vassileva, J., Gonzalez, R., & Martin, E. M. (2012). A comparison of delay discounting among substance users with and without suicide attempt history. Psychology of addictive behaviors, 26, 980-985.
Riblet, N. B., Shiner, B., Young-Xu, Y., & Watts, B. V. (2017). Strategies to prevent death by suicide: meta-analysis of randomised controlled trials. The British Journal of Psychiatry, bjp-bp.
Sadikaj, G., Moskowitz, D. S., Russell, J. J., Zuroff, D. C., & Paris, J. (2013). Quarrelsome behavior in borderline personality disorder: Influence of behavioral and affective reactivity to perceptions of others. Journal of abnormal psychology, 122, 195-207.
Szanto, K., Clark, L., Hallquist, M., Vanyukov, P., Crockett, M., & Dombrovski, A. Y. (2014). The cost of social punishment and high-lethality suicide attempts in the second half of life. Psychology and aging, 29, 84-94.