ISSN 0439-755X
CN 11-1911/B

Acta Psychologica Sinica ›› 2023, Vol. 55 ›› Issue (12): 1966-1978.doi: 10.3724/SP.J.1041.2023.01966

• Reports of Empirical Studies • Previous Articles     Next Articles

Pain avoidance and effective connectivity between dlPFC and insula in suicide attempters with major depressive disorder

HAO Ziyu, LI Huanhuan(), LIN Yixuan   

  1. Department of Psychology, Renmin University of China, Beijing 100872, China
  • Published:2023-12-25 Online:2023-10-16
  • Contact: LI Huanhuan, E-mail:


Previous research has established pain avoidance as a critical and stable variable for identifying suicide attempts across samples. However, the neural mechanisms underlying pain avoidance and its relationship to suicide attempts remain unclear. Additionally, most previous research on the brain mechanisms of suicide has focused on single brain regions or circuits, neglecting the principle that the brain is organized into networks. Psychological pain is a complex, negative emotional experience that involves cognitive appraisal, psychosomatic responses, and motivational tendencies. Therefore, the brain mechanisms underlying psychological pain mainly involve impairments in pain processing, motivation, and decision-making circuits. Therefore, a more comprehensive approach is necessary to explore the neural mechanisms underlying psychological pain and its relationship with suicide attempts.

The “pain matrix” is a functional brain network that plays an important role in pain processing. The medial pathway of the pain matrix is responsible for processing affective pain information and is considered the neural basis of psychological pain. The three-dimensional psychological pain model divides psychological pain into cognitive, affective, and motivational components. The medial pathway of the pain matrix is responsible for processing the affective components of psychological pain. Pain arousal, as a cognitive component, represents an individual's cognitive appraisal of painful past situations based on the recollection of past negative events and is closely related to the function of the hippocampus. Pain avoidance, as a motivational component, is closely related to the functions of brain regions involved in cognitive control, with the dlPFC being the core node for top-down cognitive control. Therefore, a combination of the medial pathway, dlPFC, and the hippocampus, defined as a “pain processing network”, can be used to explore the neural mechanisms of psychological pain and its relation to suicide attempts.

Previous studies have linked neural mechanisms to the pathophysiology of suicide, based on brain region-specific functional and structural alterations. However, owing to the lack of directional analyses of multiregional connectivity, little is known about the top-down or bottom-up interaction patterns of the crucial brain regions mentioned above during pain processing in suicide. Therefore, this study aims to explore directional interaction patterns of multiple brain regions in the “pain processing network” and their relationships to pain avoidance, painful feelings, and suicide attempts. This may shed new light on the suicidal brain from a psychological pain-processing perspective.

A total of 35 healthy controls (HC), 25 MDD patients with a history of suicide attempts (MDD-SA), and 48 MDD patients without such a history (MDD-NSA) participated voluntarily in this study. All participants completed measures of interested questionnaires and underwent resting-state functional magnetic resonance imaging. Granger causal analysis (GCA) was used to explore the effective connectivity patterns of brain regions in the pain-processing network.

Behavioral results showed that the total TDPPS scores [F(2, 101) = 78.80, p < 0.001, η2p = 0.76], pain arousal [F(2, 101) = 57.81, p < 0.001, η2p = 0.70], painful feelings [F(2, 101) = 61.10, p < 0.001, η2p = 0.71], pain avoidance subscale[F(2, 101) = 53.65, p < 0.001, η2p = 0.60], BDI [F(2, 101) = 136.51, p < 0.001, η2p = 0.85], BSI-C [F(2, 101) = 29.11, p < 0.001, η2p = 0.54] and BSI-W [F(2, 101) = 53.65, p < 0.001, η2p = 0.68] were significantly higher in the MDD-SA group than in the MDD-NSA and HC groups.

Additionally, as shown in Figure 1, the MDD-SA group showed weaker top-down effective connectivity from the right dorsolateral prefrontal cortex (dlPFC) to the right insula [F(2, 101) = 3.13, p = 0.018, η2p = 0.11] and stronger effective connectivity from the right lateral orbitofrontal cortex (OFC) to the left hippocampus [F(2, 101) = 3.27, p = 0.014, η2p = 0.12], and the right hippocampus to the right thalamus[F(2, 101) = 3.68, p = 0.008, η2p = 0.13], than the other two groups. Interestingly, the MDD-SA group showed stronger parallel effective connectivity from the left amygdala to the right amygdala [F(2, 101) = 2.08, p = 0.088, η2p = 0.08] than the MDD-NSA and HC groups. The MDD-NSA group showed weaker bottom-up effective connectivity from the left insula to the right dlPFC [F(2, 101) = 2.47, p = 0.050, η2p = 0.09], the left OFC to the left dlPFC [F(2, 101) = 2.65, p = 0.038, η2p = 0.10] and from the left amygdala to the left dlPFC [F(2, 101) = 3.02, p = 0.021, η2p = 0.11] than did the other two groups. The MDD-NSA group showed stronger effective connectivity from the left dlPFC to the left lateral OFC [F(2, 101) = 3.51, p = 0.010, η2p = 0.12], and left medial OFC to the right medial OFC [F(2, 101) = 3.60, p = 0.009, η2p = 0.13] than the MDD-SA and HC groups. Notably, blunted effective connectivity from the right dlPFC to the right insula was negatively correlated with pain avoidance scores, suicidal ideation at the worst time point, and the number of suicide attempts (see Figure 2).

As a pilot study focused on the directionally interactive patterns of the pain-processing network in suicide attempters, the results suggest that reduced top-down connectivity from the dlPFC to the insula plays a crucial role in the shared neural mechanisms underlying pain avoidance and suicide attempts. This blunting of top-down effective connectivity reflects the difficulty of suicide attempters in achieving effective cognitive control and adaptive behaviors in response to intense negative emotions. Consequently, the individual chooses to commit suicide, a non-adaptive behavior driven by powerful pain avoidance motivations. In contrast, the MDD-NSA group exhibited blunted bottom-up connectivity, which is distinguished from the connectivity pattern of the pain-processing network in the MDD-SA group. Therefore, this study provided neuroimaging evidence for the superiority of pain avoidance over depression in identifying suicide attempts.