Registered ReportThe role of the vestibular system in value attribution to positive and negative reinforcers
Introduction
Human decision-making processes can hardly be understood in full without taking into account one individual’s physiological state and needs (Berridge & Robinson, 2003; Craig, 2002; Critchley, Wiens, Rotshtein, Öhman, & Dolan, 2004; Damasio, 1996; Gray & Critchley, 2007; Namkung, Kim, & Sawa, 2017; Naqvi & Bechara, 2010; Naqvi, Rudrauf, Damasio, & Bechara, 2007; Paulus, 2007). “Somatic markers” originating from bioregulatory processes can bias the subjective desirability of stimuli, thus guiding cognition and behavior (Craig, 2002; Damasio, 1996; Mayer, 2011; Morton, Cummings, Baskin, Barsh, & Schwartz, 2006; Namkung et al., 2017; Paulus, 2007; Seth, 2013). For example, think how desirable would be one’s preferred food in normal conditions versus while experiencing mild nausea. A global interoceptive state is defined by the balance and integration of visceral, autonomic, somatosensory, motor and vestibular inputs, wherein the insula is known to represent a key structure in the integration of these signals (Craig, 2002; Namkung et al., 2017). The incentive value of any stimulus can be encoded in an abstract fashion, but also in relation to the expected effect on physiological homeostasis (Berridge & Robinson, 2003; Palminteri et al., 2012), before being exploited for guiding human choices via connections with other structures such as limbic areas and the dorsal striatum (Berridge & Robinson, 2003; Namkung et al., 2017; Palminteri et al., 2012). This double coding of one stimulus (i.e. abstract vs. homeostasis-related) is therefore capable to explain why the same stimulus can assume different motivational values under different interoceptive states.
One important, yet overlooked, bodily signal is the uncoupling of normally tightly linked inputs, i.e. visual and vestibular information, which in the most extreme case results in motion sickness and nausea (Kohl, 1983; Lackner, 2014; Treisman, 1977). Motion sickness is a complex syndrome which manifests itself not only with nausea and vomiting but also with a variable degree of pallor, cold sweating, drowsiness, and dizziness. Though the physiological origins of motion sickness are not yet fully understood, the most widely accepted theory posits that sensory conflicts (i.e. a discrepancy between the expected and afferent signals from the body) may provoke the extreme reaction of neural mechanisms in the brainstem and several cortical regions (Lackner, 2014), including responses of anxiety and aversive conditioning (Balaban, 2002). A partly alternative, evolutionary account posits that motion sickness may essentially be related to poisoning (Treisman, 1977). Under this evolutionary framework, visuo-vestibular mismatches are reminiscent of the subtle but informative warning signals which follow the ingestion of noxious neurotoxins (Treisman, 1977), and thus would prompt defensive reactions such as emesis. Furthermore, such mismatches would participate to the process of aversive conditioning in the sense that they would provide to an unspecialized feeder elements for avoiding potentially noxious stimuli or environments in the future (Treisman, 1977). At any rate, the experience of visuo-vestibular mismatches may be regarded as a powerful perturbation of the human interoceptive state.
Vestibular stimulation techniques–which alter the activity of an extended parieto-insular network (Lopez, Blanke, & Mast, 2012; zu Eulenburg, Caspers, Roski, & Eickhoff, 2012) – have been found to modulate affective control, mood, purchase decision-making (in terms of desirability of products) in healthy subjects (Mast, Preuss, Hartmann, & Grabherr, 2014; Preuss, Kalla, Müri, & Mast, 2017; Preuss, Hasler, & Mast, 2014a; Preuss, Mast, & Hasler, 2014b), and to alleviate manic symptoms in patients (Carmona, Holland, & Harrison, 2009; Dodson, 2004; Levine et al., 2012). For example, Preuss, Hasler, and Mast (2014a) reported that caloric vestibular stimulation can modulate the performance in a Go/No-go task exploiting emotional images as target stimuli. In their study, affective control (i.e. the proportion of hits vs. false alarms, and thus motor inhibition) for positive images was modulated by the vestibular stimulation, and could either be reduced or increased as a function of the stimulated ear. Moreover, one recent study found that vestibular stimulation through galvanic current (Galvanic Vestibular Stimulation, GVS) decreases sensitivity to monetary rewards (Blini, Tilikete, Farnè, & Hadj-Bouziane, 2018a). Therefore, stimulating the vestibular sense may be a key and convenient mean to perturb interoceptive states or otherwise compensate for a system that does not link efficiently visceral states to optimal decision-making strategies, as for instance in addicted individuals.
In addiction disorders, cues associated with one’s behavioral addiction exert a strong attentional capture (Field, Munafò, & Franken, 2009). The magnitude of attentional bias has been found to predict relapse from treatment (Field & Cox, 2008; Garland, Franken, & Howard, 2012; Marissen et al., 2006), and has been causally related to craving (Field & Eastwood, 2005). Such cues activate automatic representations of value which can interfere with the task at hand, when they should rather be inhibited because distracting (Carpenter, Schreiber, Church, & McDowell, 2006; Cox, Fadardi, & Pothos, 2006; Gross, Jarvik, & Rosenblatt, 1993; Nijs, Franken, & Muris, 2010; Sokhadze, Stewart, Hollifield, & Tasman, 2008). For example, in the manifold variants of the Addiction–Stroop test (Cox et al., 2006), participants with behavioral addictions are typically found to be slower than controls in reading aloud the color of words related to the substance of abuse. Thus, interference costs can provide valuable clinical information, on one hand, and reducing them could represent a major therapeutic objective, on the other hand. This appears particularly appropriate as prominent models of behavioral addictions emphasize–among other societal, neurobiological, and cognitive aspects–the role of both “driving” (enhanced sensitivity to reward) and control (reduced inhibitory control) aspects, and their interaction (Baler & Volkow, 2006; Della Libera et al., 2019; Goldstein & Volkow, 2011).
Emerging evidence also points toward an important role of defective interoceptive processing in craving and in the maintenance of behavioral addictions (Gray & Critchley, 2007; Paulus, Stewart, & Haase, 2013; Verdejo-Garcia, Clark, & Dunn, 2012). New avenues for the clinical management of these patients include treatment options based on interoceptive processing (Paulus, 2007; Paulus et al., 2013), their effectiveness being explained in light of a possible modulation of insular activity. Indeed, inactivation of the insula in amphetamine-experienced rats may prevent their drug-seeking behavior and blunt the malaise associated with craving (Contreras, Ceric, & Torrealba, 2007); in humans, damage to or functional disruption of the insula may alleviate addiction to nicotine (Dinur-Klein et al., 2014; Naqvi et al., 2007). Thus, manipulating this circuitry represents an appealing way to tackle the addiction loop in the brain (Dinur-Klein et al., 2014; Droutman, Read, & Bechara, 2015; Naqvi & Bechara, 2010; Naqvi et al., 2007; Paulus, 2007). However, the effects of vestibular stimulation–which also taps onto an extended parieto-insular network (Lopez et al., 2012; zu Eulenburg et al., 2012) – have been seldom studied.
Detrimental effects on performance and control induced by value-associations have also been described in healthy individuals, and characterized as being a function of the distractors’ salience and the degree of automaticity of their processing (Bourgeois, Chelazzi, & Vuilleumier, 2016; Chelazzi, Perlato, Santandrea, & Della Libera, 2013; Failing & Theeuwes, 2018; Krebs, Boehler, & Woldorff, 2010). This effect is important because it provides a working model to study a key feature of addiction disorders in a laboratory setting. In such experimental conditions, value-associations are typically established using monetary rewards. With this study, we therefore sought to test, in healthy subjects, whether a vestibular/interoceptive perturbation is capable to decrease the unduly interference induced by intrinsically salient features (Krebs et al., 2010). Second, we sought to test whether such artificially biased interoceptive state differentially affects rewards and punishments. Whether negative reinforcers are devaluated following visuo-vestibular mismatches, as are positive ones, is currently unknown. One study found that inactivation of the posterior insula in rats may result in disrupted acquisition of both conditioned place preference and place avoidance (Li, Zhu, Meng, Li, & Sui, 2013). Indeed, a perturbed system that strives for homeostasis may bring about a general blunted response to external stimuli, as physiologic balance would become the main priority; this predicts decreased sensitivity to punishments (Scenario 1 in Fig. 2C). However, the same study also found evidence for a neural dissociation: lesions to the anterior insula selectively disrupted conditioned place avoidance, but not conditioned preference (Droutman et al., 2015; Li et al., 2013). One possibility is thus to observe a similar behavioral specificity. For example, in the study of Preuss, Hasler, and Mast (2014a), affective control was modulated for positive, but not negative images. Should the vestibular perturbation be specialized for appetitive stimuli, negative reinforcers may be unaffected (Scenario 2 in Fig. 2C). However, differently from positive reinforcers, punishments have a threatening nature and negative valence, which appear to match the nature of the altered interoceptive processing and may resonate with it. As visuo-vestibular mismatches may be exploited as (further) warning signal (e.g. signaling the contact with neurotoxins, Treisman, 1977), negative reinforcers may become more salient and, ultimately, more effective in capturing attention. We therefore put forward the possibility that the effects of negative reinforcers may be enhanced by a vestibular stimulation (Scenario 3 in Fig. 2C, and possible neural substrates in paragraph 1.5). In the context of addiction disorders, this would be equally desirable to a decreased sensitivity to positive rewards to the extent that, in the context of decision making, short and long term negative consequences are weighted comparatively more than short term positive effects, i.e. “myopia” for the future is counteracted (Bechara, Dolan, & Hindes, 2002).
We administered GVS to healthy participants engaged in a task evoking conflict and the need to inhibit irrelevant information (an ad-hoc adaptation of Eriksen & Eriksen, 1974; Eriksen, 1995). Participants received, in three different days, one sham stimulation and two active GVS stimulations, with opposite hemispheric lateralization (Blini, Tilikete, et al., 2018a). We used a modified version of the Flankers Task (FT), whereby five arrows were presented on screen: the task consisted in indicating in which direction the central arrow was pointing. The four flanking arrows could either point toward the same (congruent) or the opposite direction (incongruent), thereby creating interference costs in the incongruent condition. Depending on their color, flankers could also signal the possibility of receiving monetary incentives or losses, as a function of the subjects’ performance. Interference costs were expected to increase when–as compared to a Neutral (N) condition without reward–Potential Gains (PG) or Potential Losses (PL) were at play (Krebs et al., 2010), reflecting increased attentional processing. We had two predictions: 1) interference costs would be maximal, in PG conditions, for the sham stimulation, but reduced by GVS, especially with a Right-Anodal montage (which was associated with the greatest reduction in sensitivity to monetary rewards in a previous study, Blini, Tilikete, et al., 2018a); 2) interference costs for PL, on the contrary, would be enhanced with either GVS condition with respect to the sham stimulation (Scenario 3 in Fig. 2C).
In Fig. 1, a schematic depiction of a neural mechanism possibly involved in the link between vestibular perturbations and motivation is presented, largely inspired by the Embodied Predictive Interoception Coding (EPIC) model of Barrett and Simmons (2015). The EPIC model stems from recent active inference accounts (e.g. Adams, Shipp, & Friston, 2013), holding that perception is a process of statistical inference and prediction about the causes of sensations (also see Seth, 2013). Under this framework, the (external) context at hand would be first evaluated through a prior model, based on past experience, about the most likely output under the circumstances. This comparison allows the brain to issue timely predictions about the most appropriate bodily response, the required allostatic adjustments, and the interoceptive sensations that are expected to arise. These predictions are thus the basis for the perception of and the action toward the external context, which can modify the context in turn (e.g. by modulating attentional resources, or by prompting approach-withdraw behaviors). A widespread network of visceromotor cortices–e.g. posterior orbitofrontal cortex, posterior ventral medial prefrontal cortex, anterior insula, and anterior cingulate cortex–has been putatively ascribed to this function: these areas (agranular cortices) present a laminar architecture characterized by a large number of pyramidal neurons in the deepest layers in face of fewer projection neurons in the upper layers, a feature that suggests a role in sending sensory predictions to granular cortices, which are better equipped for computing prediction errors. Indeed, the “primary interoceptive cortex”, i.e. the granular posterior insula, appears as one key region in which these predictions are compared to the actual incoming sensations. If predictions are roughly accurate, the difference will be null or negligible; if large discrepancies between predictions and actual sensory signals are, instead, detected, one prediction error signal will back-propagate to visceromotor cortices and modulate their activity. As a result, the prediction error may adjust predictions and the prior model that generated them directly, in both cases affecting the actions the body will undertake and the evaluation of the context.
Under this framework, we speculated that, while the sham stimulation would not (sensibly) interfere with the process of sensory integration, vestibular stimulation would bias the incoming sensory signals toward a negative and unpleasant bodily state, in light of visuo-vestibular conflict. As a result, prediction errors would be larger than what expected on the basis of previous experience, and will guide the subsequent predictions toward a more negative valence attribution. The responses of the brain may therefore adapt differently to the context at hand. Although the present study did not envisage neural measures able to test this mechanistic model specifically, we predicted that a positive context, such that arising from monetary rewards, would be resized and devaluated (Blini, Tilikete, et al., 2018a); a negative context, arising when potential losses are at stake, would be magnified instead. This further motivated our prediction (Scenario 3 in Fig. 2C) of enhanced interference costs for PL in face of reduced interference costs for PG with concurrent GVS.
Section snippets
Methods
The registered protocol (https://osf.io/2htd3) and all supporting materials (Supplementary Materials: https://osf.io/b9ezq/) are available on the Open Science Framework website.
Subjective visual vertical (SVV)
When assessing the SVV, we found, as predicted, a large effect of Starting Side (χ2(1) = 66, pfdr< .001, ηp2 = .59), lines originally tilted in the clockwise direction being associated with a more pronounced clockwise bias. Additionally, the main effect of GVS was large and significant (χ2(2) = 62.76, pfdr< .001, ηp2 = .42). Post-hoc t-tests showed that all GVS conditions differed: Left-Anodal was associated with a bias in the counter-clockwise sense with respect to the sham (β = .166,
Motivation, valence and conflict
Motivation–experimentally provided, for example, in the form of monetary rewards–can act by invigorating one’s action or willingness to exert an effort (Chong, Bonnelle, & Husain, 2016; Chong et al., 2015; Muhammed et al., 2016), or by sharpening cognitive processes such as attention and memory (Abe et al., 2011; Anderson, Laurent, & Yantis, 2011; Chelazzi et al., 2013; Della Libera & Chelazzi, 2006, 2009; Engelmann & Pessoa, 2007; Hickey, Chelazzi, & Theeuwes, 2010). However, recent
Conclusions
In this paper, we report an asymmetry in the processing of gains and losses, with the latter seemingly more effective in capturing human attention. When negative reinforcers are attached to distractors, these stimuli appear to be processed more thoroughly, and the interference costs arising in presence of conflicting information are consequently enhanced.
We also report some evidence for Right-Anodal galvanic vestibular stimulation to further increase the salience of losses, as suggested by even
Author contributions
Elvio Blini: Conceptualization; Data curation; Formal analysis; Funding acquisition; Investigation; Methodology; Project administration; Resources; Software; Validation; Visualization; Original draft; Review & editing.
Caroline Tilikete: Conceptualization; Methodology; Project administration; Supervision; Validation; Review & editing.
Leonardo Chelazzi: Conceptualization; Methodology; Supervision; Validation; Review & editing.
Alessandro Farné: Conceptualization; Methodology; Project
Open Practices
The study in this article earned Open Materials, Open Data and Preregistered badges for transparent practices. Materials and data for the study are available at https://osf.io/b9ezq/.
Acknowledgments
EB was supported by: the European Union’s Horizon 2020 research and innovation programme (Marie Curie Actions) under grant agreement MSCA-IF-2016-746154; a grant from MIUR (Departments of Excellence DM 11/05/2017 n. 262) to the Department of General Psychology, University of Padova. AF is supported by the James S. McDonnell Scholar award. FHB received funding from the French National Research Agency (ANR-14-CE13-0005-1). The study was performed within the framework of the LABEX CORTEX
References (124)
- et al.
Reward improves long-term retention of a motor memory through induction of offline memory gains
Current Biology
(2011) - et al.
Mixed-effects modeling with crossed random effects for subjects and items
Journal of Memory and Language
(2008) Neural substrates linking balance control and anxiety
Physiology & Behavior
(2002)- et al.
Drug addiction: The neurobiology of disrupted self-control
Trends in Molecular Medicine
(2006) - et al.
Random effects structure for confirmatory hypothesis testing: Keep it maximal
Journal of Memory and Language
(2013) - et al.
Decision-making and addiction (part II): Myopia for the future or hypersensitivity to reward?
Neuropsychologia
(2002) - et al.
Parsing reward
Trends in Neurosciences
(2003) - et al.
Probing the role of the vestibular system in motivation and reward-based attention
Cortex; a Journal Devoted To the Study of the Nervous System and Behavior
(2018) - et al.
Conflict monitoring and anterior cingulate cortex: An update
Trends in Cognitive Sciences
(2004) - et al.
Chapter 14—how motivation and reward learning modulate selective attention
Drug Stroop performance: Relationships with primary substance of use and treatment outcome in a drug-dependent outpatient sample
Addictive Behaviors
Rewards teach visual selective attention
Vision Research
Chapter 4—quantifying motivation with effort-based decision-making paradigms in health and disease
Dopamine enhances willingness to exert effort for reward in Parkinson’s disease
Cortex; a Journal Devoted To the Study of the Nervous System and Behavior
Smoking cessation induced by deep repetitive transcranial magnetic stimulation of the prefrontal and insular cortices: A prospective, randomized controlled trial
Biological Psychiatry
Revisiting the role of the insula in addiction
Trends in Cognitive Sciences
Attentional bias in addictive behaviors: A review of its development, causes, and consequences
Drug and Alcohol Dependence
Interoceptive basis to craving
Neuron
Emotional foundations of cognitive control
Trends in Cognitive Sciences
The influence of reward associations on conflict processing in the Stroop task
Cognition
Beneficial effects of caloric vestibular stimulation on denial of illness and manic delusions in schizoaffective disorder: A case report
Brain Stimulation
Common and distinct networks underlying reward valence and processing stages: A meta-analysis of functional neuroimaging studies
Neuroscience and Biobehavioral Reviews
The human vestibular cortex revealed by coordinate-based activation likelihood estimation meta-analysis
Neuroscience
Balancing Type I error and power in linear mixed models
Journal of Memory and Language
The insula: An underestimated brain area in clinical neuroscience, psychiatry, and neurology
Trends in Neurosciences
Food-related Stroop interference in obese and normal-weight individuals: Behavioral and electrophysiological indices
Eating Behaviors
Critical roles for anterior insula and dorsal striatum in punishment-based avoidance learning
Neuron
Caloric vestibular stimulation modulates affective control and mood
Brain Stimulation
Predictions not commands: Active inference in the motor system
Brain Structure & Function
Antero-posterior versus lateral vestibular input processing in human visual cortex
Open science challenges, benefits and tips in early career and beyond
PLOS Biology
Value-driven attentional capture
Proceedings of the National Academy of Sciences
Interoceptive predictions in the brain
Nature Reviews Neuroscience
Parsimonious mixed models
Fitting linear mixed-effects models using lme 4
Controlling the false discovery rate: A practical and powerful approach to multiple testing
Journal of the Royal Statistical Society. Series B (Methodological)
Approaching the bad and avoiding the good: Lateral prefrontal cortical asymmetry distinguishes between action and valence
Journal of Cognitive Neuroscience
Mind the depth: Visual perception of shapes is better in peripersonal space
Psychological Science
Ipsilesional impairments of visual awareness after right-hemispheric stroke
Frontiers in Psychology
Conflict monitoring and decision making: Reconciling two perspectives on anterior cingulate function
Cognitive, Affective & Behavioral Neuroscience
Dorsal anterior cingulate cortex: A role in reward-based decision making
Proceedings of the National Academy of Sciences
Functional connectivity of reward processing in the brain
Frontiers in Human Neuroscience
Hemispheric asymmetry for emotional stimuli detected with fMRI
Neuroreport
Extending the functional cerebral systems theory of emotion to the vestibular modality: A systematic and integrative approach
Psychological Bulletin
Are losses more effective than rewards in improving performance in a cognitive task?
Motivation Science
Anterior cingulate cortex, error detection, and the online monitoring of performance
Science
Inactivation of the interoceptive insula disrupts drug craving and malaise induced by lithium
Science
The addiction-stroop test: Theoretical considerations and procedural recommendations
Psychological Bulletin
How do you feel? Interoception: The sense of the physiological condition of the body
Nature Reviews Neuroscience
Cited by (3)
Effects of attentional shifts along the vertical axis on number processing: An eye-tracking study with optokinetic stimulation
2022, CognitionCitation Excerpt :This means that in different analyses the random slopes could differ, i.e., in one set of analyses a random slope could improve the model when compared to a model with no random slopes, while in another set of analyses a random slope might not significantly ameliorate the model and therefore it would be not retained in the final model. For a more detailed description of this pipeline, see (Blini, Tilikete, Chelazzi, Farnè, & Hadj-Bouziane, 2020; Blini, Tilikete, Farnè, & Hadj-Bouziane, 2018). In this phase, models presenting fitting problems (e.g., failure in convergence) were systematically excluded, as to avoid overfitting.
Pupil size as a robust marker of attentional bias toward nicotine-related stimuli in smokers
2023, Psychonomic Bulletin and Review