Attenuation of access to internal states in high obsessive-compulsive individuals might increase susceptibility to false feedback: Evidence from a visuo-motor hand-reaching task

Highlights

• Influence of false feedback in low vs. high obsessive compulsive (OC) participants.

• Hand reaching accuracy was tested after training with valid and rotated feedback.

• Reaching accuracy of High OC had larger decrease after training with false feedback.

• Attenuated access to internal states in OC individuals may explain findings.

Abstract

Background and objectives

The Seeking Proxies for Internal States (SPIS) model of obsessive-compulsive disorder (OCD) posits that obsessive-compulsive (OC) individuals have attenuated access to their internal states. Hence, they seek and rely on proxies, or discernible substitutes for these internal states. In previous studies, participants with high OC tendencies and OCD patients, compared to controls, showed increased reliance on external proxies and were more influenced by false feedback when judging their internal states. This study is the first to examine the effects of false feedback on performance of hand movements in participants with high and low OC tendencies.

Method

Thirty-four participants with high OC tendencies and 34 participants with low OC tendencies were asked to perform accurate hand reaches without visual feedback in two separate sessions of a computerized hand-reaching task: once after valid feedback training of their hand location and once with false-rotated feedback. We assessed the accuracy and directional adaptation of participants’ reaches.

Results

As predicted, high OC participants evidenced a larger decrease in their hand positioning accuracy after training with false feedback compared to low OC participants.

Limitations

The generalization of our findings to OCD requires replication with a clinical sample.

Conclusions

These results suggest that in addition to self-perceptions, motor performance of OC individuals is prone to be overly influenced by false feedback, possibly due to attenuated access to proprioceptive cues. These findings may be particularly relevant to understanding the distorted sense of agency in OCD.

Introduction

A recent comprehensive model of obsessive-compulsive disorder (OCD) postulates that a core feature of the disorder is impaired access to internal states, including emotions and preferences as well as bodily states and sensations. According to the Seeking Proxies for Internal States model (SPIS; Lazarov, Dar, Oded, & Liberman, 2010; Liberman & Dar, 2009), obsessive-compulsive (OC) individuals seek and resort to proxies to compensate for the attenuated access to their internal states. Proxies are substitutes for the internal state that the individual perceives as easier to monitor and discern, such as rules, behaviors, or environmental stimuli (Lazarov, Liberman, Hermesh, & Dar, 2014; Liberman & Dar, 2009). According to the SPIS model, the attenuation of access to internal states in OCD can account for the complex phenomenology of the disorder, including prevailing doubts, repetitive checking, behavior governed by rules and rituals, and a distortion in the sense of agency.

Previous studies have supported the SPIS model by showing that individuals with high OC tendencies have reduced access to various internal states, including relaxation level (Lazarov et al., 2010), muscle tension (Lazarov, Dar, Liberman, & Oded, 2012a,b, Zhang et al., 2017 and Lazarov et al., 2014 in OCD patients), time perception (Gilaie-Dotan, Ashkenazi, & Dar, 2016), sense of understanding (Dar et al., submitted), movement initiation (Ezrati, Sherman, & Dar, 2018) and emotions (Dar, Lazarov, & Liberman, 2016). In various tasks involving these internal states, high OC participants and OCD patients appeared to have attenuated access to these states, were more inclined to seek and depend on external proxies for them, and were more vulnerable to the effects of false feedback. For example, using a muscle tensing task, Lazarov et al. (2014) demonstrated that OCD patients had larger errors when asked to reproduce specific degrees of muscle tension, as compared with non-clinical and anxiety control participants. This difference was eliminated when participants were provided with biofeedback as an external proxy for this internal state. In another study (Lazarov, Dar, Liberman, & Oded, 2012b), high and low OC participants were instructed to relax their forearm muscles while viewing false preprogrammed “feedback” on their muscle tension. Each participant underwent two successive phases of putative feedback, one indicating a gradual increase in muscle tension and one indicating a gradual decrease in muscle tension. Following each phase, participants rated their perceived level of muscle tension. As predicted, the ratings of the high OC participants indicated that they relied more on the (false) biofeedback proxy in judging their own level of muscle tension.

Though not originally described as such, the aforementioned studies involving muscle tension suggest that OCD tendencies may be related to deficient proprioception. Proprioception, derived from the Latin word proprius meaning “one's own,” is defined as information derived from skin and musculoskeletal receptors especially on body movement, body position, muscle tension, balance and sense of effort (Proske & Gandevia, 2012). Proprioception is also considered to be a precursor of the sense of agency (SoA), which is “the registration that I am the initiator of my actions” (Synofzik, Vosgerau, & Voss, 2013). Agency experiences rely on integration and compatibility of predictive and postdictive cues (e.g., Gentsch & Synofzik, 2014; Synofzik et al., 2013; Synofzik, Vosgerau, & Lindner, 2009). In the case of movement, predictive cues include the internal copy of the motor command (Synofzik et al., 2013, 2009; Tsakiris, Prabhu, & Haggard, 2006). Postdictive cues may include external cues (e.g., seeing one's hand moving) and internal cues of proprioception (e.g., feeling the position and movement from skin, muscle and joint receptors). Successful integration of these cues and an intention-sensory outcome compatibility is believed to lead to the SoA (Gentsch & Synofzik, 2014).

Convergent experimental data from a variety of paradigms suggest a reduced SoA in OC individuals (Belayachi & Van der Linden, 2010; Gentsch, Endrass, & Kathmann, 2012; Oren, Eitam, & Dar, 2017, 2016; Rossi et al., 2005). More pertinent to the current study, recent findings comparing active vs. passive acquisition in a head repositioning task suggest that high OC individuals have deficient access to internal cues involved in active movement (Ezrati et al., 2018). Attenuated access to proprioceptive signals may be one of the factors that account for this finding, and in general for the reduced SoA of people with OC tendencies and OCD.

Tasks requiring accurate hand movements are particularly promising for enriching our knowledge on proprioception in OCD. This is because the performance of accurate hand movements demands integration of different types of cues, from an efferent copy of the intention to act, through internal proprioceptive information from receptors in the muscles, joints and skin, to external feedback from exteroceptors, such as the eyes (Proske & Gandevia, 2012).

Studies using false feedback on muscle tension indicate that OC individuals are prone to form their assessments of their own muscle tension in accord with the false feedback they have received (Lazarov et al., 2012a; 2012b, 2014). However, these findings are not informative regarding any potential functional impact of the false feedback; specifically, would high OC individuals be more affected by false feedback not just in their assessment of internal states but on subsequent performance that relies on accurate assessment of these internal states? Accordingly, the current study had two aims. First, to extend the predictions of the SPIS model to the domain of proprioception; and second, to evaluate the impact of false feedback on subsequent performance in individuals with high OC tendencies.

We used a point-to-point hand-positioning task (similar to the task used in Inzelberg, Flash, Schechtman, & Korczyn, 1995 and Jones, Cressman & Henriques, 2009). Participants performed the movements on two separate occasions, each with a different feedback condition. In the valid feedback session, they received visual feedback that was aligned with their hand location, whereas in the false feedback session they received rotated visual feedback. When participants train with visual feedback on their hand location that mismatches their proprioceptive sensation, they tend to rely more on the visual feedback than on the proprioceptive information, at least when the extent of the mismatch is relatively small (Jones, Cressman & Henriques, 2009). Under such conditions, subjects correct their movements based on visual feedback, such that a new mapping between visual input and motor output is learned (i.e. visuomotor adaptation; Krakauer, Ghilardi, & Ghez, 1999; Cressman & Henriques, 2009). Following the previous training, when participants perform reaches without visual feedback, they continue, for a certain period, to produce movements in the direction opposite to the visual rotation. These movements are referred to as “reach after-effects" and result in errors in the direction of movement and final position of the hand (Martin et al., 1996; Krakauer et al., 1999; Krakauer, Pine, Ghilardi, & Ghez, 2000; Baraduc & Wolpert, 2002).

Following the SPIS model, we predicted that compared to individuals with low OC tendencies, individuals with high OC tendencies will rely more on visual feedback as an external proxy and less on internal proprioceptive information when performing point-to-point movements. Therefore, their performance in the “after-effects” blocks should be more affected by the absence of external visual information. Following the same line of thought, we expected that supplying false feedback training will have a larger effect on the performance of high OC participants, as compared to low OC participants.