Elsevier

NeuroImage

Volume 63, Issue 2, 1 November 2012, Pages 882-893
NeuroImage

Neural correlates of causality judgment in physical and social context—The reversed effects of space and time

https://doi.org/10.1016/j.neuroimage.2012.07.028Get rights and content

Abstract

The perception of causal relationships is crucial to understanding and interacting with our physical and social environment. However, whether the same or different neural processes are involved in perceiving physical and social causality is unknown. Therefore, this study is focused on commonalities and differences in the neural correlates of causality perception in both contexts.

During fMRI data-acquisition, participants judged causal relationships of objects in two types of animated video clips (physical/social) with similar manipulations of temporal and spatial stimulus characteristics. Four conditions were analyzed in a two-factorial design [physical causal (PC), physical non-causal (PNC), social causal (SC), social non-causal (SNC)].

We found that higher angles and longer time delays led to decreasing judgments of causality in the physical context, whereas the same manipulations led to increasing judgments in the social context.

Instead of a common network for causal judgments (PC > PNC  SC > SNC), we found a reversed activation pattern for the factors context and judgment. PC and SNC [(PC > PNC) > (SC > SNC)] produced activations in the bilateral insula, the right angular and inferior frontal gyrus and the medial supplementary motor area. PNC and SC [(PC > PNC) < (SC > SNC)] produced activity in medial frontal, left superior temporal and anterior cingulate brain regions.

Our data suggest, that the same brain regions contribute to the impression of physical and social causality. However, they demonstrate a reversed activation pattern that reflects the stimulus characteristics of the respective conditions. Thus, specific stimulus characteristics are crucial for the perception of causality.

Highlights

► Causal relationships are not restricted to objects and physical forces. ► Spatial and temporal manipulations are crucial for the perception of causality. ► Space and time have reversed effects on social vs. physical causality judgments. ► No common network for causal judgments in social in contrast to physical context ► Reversed activation pattern for the factors context and judgment.

Introduction

The ability to detect causal relationships in everyday life is of fundamental importance. Knowledge about the processes in the human brain that lead to causal impressions may be helpful for a better understanding of aberrant inferences of causality in mental disorders like autism (Congiu et al., 2010, Klin and Jones, 2008, Saygin et al., 2010) and schizophrenia (Brune, 2005, Tschacher and Kupper, 2006).

The most common phenomenon investigated in the research on the perception of physical causality in the past decades is probably the "launching effect" (Michotte, 1946/1963, Wagemans et al., 2006). When an object A moves towards a stationary object B, and after a contact B moves on, most people describe this event as causal: A is the reason that B moved. Many variations of this kind of experiment have been used to investigate the key parameters that lead to the perception of causality. Those key parameters include a gap between both objects (Oakes and Kannass, 1999, Saxe and Carey, 2006, Schlottmann and Anderson, 1993, Yela, 1952), time delays before the second object starts to move (Guski and Troje, 2003, Schlottmann and Shanks, 1992, Schlottmann et al., 2006, Young and Sutherland, 2009), and the trajectory of movement (Straube and Chatterjee, 2010, Straube et al., 2011) These studies generally suggest that spatial and temporal manipulations play an important role in this context.

Given that spatial and temporal parameters affect causal inferences in physical contexts, the question arises if they contribute similarly to causal judgments in social contexts. The recognition of social interactions is a highly complex process. To investigate social interactions experimentally, it is necessary to minimize the influencing variables in the stimulus material in order to isolate the social processing (Adolphs, 2010). Heider and Simmel famously showed that, under certain conditions, participants bestow human attributes to moving simple geometric forms (Abell et al., 2000, Castelli et al., 2000, Heider and Simmel, 1944, Rimé et al., 1985, Tremoulet and Feldman, 2006). For this attribution to occur, again spatial and temporal manipulations play a crucial role (Carrozzo et al., 2010, Falmier and Young, 2008, Scholl and Tremoulet, 2000, Tremoulet and Feldman, 2000, Tremoulet and Feldman, 2006). In addition, instructions, prior experience and knowledge of the participants have a strong effect on their ratings (Gemelli and Cappellini, 1958, Gruber et al., 1957, Powesland, 1959, Schlottmann et al., 2006).

More recent studies also use imaging methods like functional magnetic resonance imaging (fMRI) or positron emission tomography (PET) to characterize the neural correlates of causality. In the physical context, Blakemore et al. found activations in the middle temporal gyrus and the right intraparietal sulcus with launching events (Blakemore et al., 2003). For causal events in contrast to non-causal events they found activations in both medial and superior temporal areas and the left intraparietal sulcus (Blakemore et al., 2001). These findings support the assumption that higher-order perceptual brain regions play a role in the detection of causal relationships. Another study in this context showed activations in both superior frontal gyri and the orbitofrontal gyrus during judgments of causality (Fonlupt, 2003). Fugelsang et al. (2005) also found activations in the right superior and middle frontal cortices and in the right inferior parietal lobule. They suggest that the detection of causality is especially instantiated in the right hemisphere. Straube and Chatterjee (2010) did not find brain activations for causal in contrast to non-causal judged stimuli. However, individuals with a higher sensitivity for spatial properties in the causality judgment task had increased brain activations in the parietal lobe, whereas individual biased to use time information had increased activation in the left basal ganglia. Inconsistencies between studies as well as evidence for individual differences, could suggest a strong influence of top-down processes, which might depend on the conditions of the task and individual differences in sensitivity to, preference for or experience with a respective parameter, stimulus, or task.

In contrast to perceptual causality and the perception of animacy (e.g., Blakemore et al., 2001, Blakemore et al., 2003, Schlottmann et al., 2006) relatively few fMRI studies have investigated social causality. In many cases this is because different research approaches and alternating terms and definitions of causality are used in a social context (Gao and Scholl, 2011). It has been suggested, that the medial frontal cortex plays an important role to determine future behavior of others (Amodio and Frith, 2006). Tavares et al. showed widespread brain activations for highly animated rated stimuli, including the bilateral frontal gyrus, the right superior temporal sulcus and amygdala (Tavares et al., 2008). They concluded that the perception of social causality does not have to be triggered by perceptual systems.

Santos et al. (2010) used two simple 3D spheres to construct short videos that showed different moving patterns of the spheres (e.g. whether the moving sphere 1 did approach sphere 2 or not (“approach”); whether sphere 2 moved toward the moving sphere 1 (“responsiveness”)). The participants had to judge a stimulus as "physical", "rather physical", "rather personal" or "personal". The results revealed significant main effects of approach and responsiveness concerning the judgment for a "personal" attribution of an object.

The neural correlates of experiencing animacy according to Santos et al. were activity in the bilateral insula, extending into the superior temporal gyrus and sulcus, the medial orbitofrontal cortex, extending into the anterior cingulate cortex, and left parahippocampal gyrus extending into the left fusiform gyrus. They concluded that animacy experiences activate this "social neural network" (SNN) (Brothers, 1990). However, since the stimuli were judged on the same scale (physical–personal), this study concentrated on brain activations for increasing or decreasing the experience of animacy and did not reveal specific information about the neural substrate of causality judgments in physical as opposed to social contexts.

In the present study we aimed to investigate the neural correlates of the judgment of causality in a physical and social context. Social causality in context of this study refers to the perception of a social interaction in which one person is perceived to influence the behavior of another person. The two persons were illustrated by neutral objects in this study to keep the social context comparable to the physical context. We compared brain activations of physical and social contexts as well as activations concerning causality judgments. Both contexts used highly comparable stimuli, each including a blue and a red simple object (circle) (see Straube and Chatterjee, 2010, Straube et al., 2011). In the physical context the blue ball collided with the red ball, whereas in the social context the blue ball passed the red ball. Two instructions were used to set up the context, which differed minimally in the stimulus arrangement (see Fig. 1). The participants judged each stimulus event as causal or non-causal, while time delay and angle in movement direction were identically manipulated in the physical and social context. The experiment used many stimuli that varied the perceptual cues parametrically to account for individual differences in perceptual sensitivity and total amount of causality judgments.

Based on previous literature (e.g. Schlottmann et al., 2006) we expected opposite effects of our stimulus manipulations on physical and social causality: Small deviations in the stimulus (small angle and short time delay) lead to causal judgment in the physical task, and non-causal judgment in the social task, whereas large deviations (great angle and long time delay) lead to the opposite judgments. In the fMRI analyses, we expect to find different brain activations during physical and social tasks in brain regions predominantly related to perceptual (e.g., for physical task and spatial effects in the parietal lobe, and temporal effects in the supplementary motor area (SMA) and the basal ganglia (Straube and Chatterjee, 2010, Straube et al., 2011)) and inferential (e.g., for social tasks in the prefrontal cortex (Tavares et al., 2008)) processing mechanisms. The medial prefrontal cortex (MPFC) has previously been described as instantiating several psychological mechanisms (e.g. self-concept, mentalizing and emotional experience) and might more generally compute inexact and internally generated estimates for social phenomena (Mitchell, 2009).

We tested the following alternate hypothesis: firstly, if the perception of causality is a universal function independent of context, we expect the same brain regions to be activated for causal in contrast to non-causal trials in both the physical and the social contexts. However, if the judgment of causality is instantiated through specific context dependent neural networks, we expect significant interaction effects in brain regions previously reported with regard to perception of physical and social causality or animacy (e.g., the bilateral insula, the superior temporal gyrus/sulcus (STG/STS), the medial orbitofrontal cortex, the anterior cingulate cortex (ACC), and left parahippocampal gyrus (Santos et al., 2010), right medial frontal gyrus and in the inferior parietal lobule (Fugelsang et al., 2005); bilateral superior frontal gyri and the orbitofrontal gyrus (Fonlupt, 2003); medial and superior temporal areas and the bilateral intraparietal sulcus (Blakemore et al., 2001, Blakemore et al., 2003)). Such an interaction of context and judgment would further suggest that brain activity depends on stimulus parameters rather than on judgments, given the expected opposite effects of time and space in the physical and social causality contexts.

Section snippets

Participants

20 subjects were included in the analyses (10 male/female, mean age = 24.3 years). All were right-handed and had a normal or corrected-to-normal vision. None of them reported significant medical treatment because of neurological or psychiatric reasons.

Initially, twenty-two participants took part in the study. However, one male participant was excluded from the fMRI-analysis because of excessive head movement and one female participant was excluded because of technical problems during the

Behavioral results

The participants rated 42.86% (SD = 12.36%, range = 24.49–69.39%) of the physical stimuli and 56.31% (SD = 17.68%, range = 28.57–88.78%) of the social stimuli as causal. Thus, there is a significant difference in the degree of detected causality in both conditions (t(19) =  2.411, p < 0.026). Furthermore, correlation analyses indicate no significant correlation between causality judgments in both contexts (Correlation coefficient =  0.361, p > 0.118).

With a greater angle deviations and time delays the

Discussion

Causal relationships are not restricted to objects and physical forces. They are also inferred in social situations. In the present study we aimed to compare the neural correlates of the perception of causality in physical and social contexts using similar stimuli and manipulations of spatial and temporal parameters.

Behaviorally, we found that spatial and temporal event parameters were important for causality judgments in both contexts. However, whereas violations of spatial continuity and

Acknowledgments

Supported by a Research Grant of the University Medical Center Giessen and Marburg (UKGM; Project number: 10/2010MR). BS is supported by the BMBF (BMBF; project number: 01GV0615). We thank PhD Adam J. Woods for his comments and corrections.

References (66)

  • D. Kawawaki et al.

    Anterior and superior lateral occipito-temporal cortex responsible for target motion prediction during overt and covert visual pursuit

    Neurosci. Res.

    (2006)
  • M. Keehner et al.

    Modulation of neural activity by angle of rotation during imagined spatial transformations

    NeuroImage

    (2006)
  • J.S. Mayer et al.

    Common neural substrates for visual working memory and attention

    NeuroImage

    (2007)
  • J.P. Mitchell

    Social psychology as a natural kind

    Trends Cogn. Sci.

    (2009)
  • A.M. Morcom et al.

    Does the brain have a baseline? Why we should be resisting a rest

    NeuroImage

    (2007)
  • T. Nichols et al.

    Valid conjunction inference with the minimum statistic

    NeuroImage

    (2005)
  • N.S. Santos et al.

    Animated brain: a functional neuroimaging study on animacy experience

    NeuroImage

    (2010)
  • R. Saxe et al.

    The perception of causality in infancy

    Acta Psychol. (Amst)

    (2006)
  • A. Schlottmann

    Is perception of causality modular?

    Trends Cogn. Sci.

    (2000)
  • A. Schlottmann et al.

    Perceived physical and social causality in animated motions: spontaneous reports and ratings

    Acta Psychol.

    (2006)
  • B.J. Scholl et al.

    Perceptual causality and animacy

    Trends Cogn. Sci.

    (2000)
  • J. Schultz et al.

    Activation in posterior superior temporal sulcus parallels parameter inducing the percept of animacy

    Neuron

    (2005)
  • M.U. Shad et al.

    Neurobiology of decision-making in adolescents

    Behav. Brain Res.

    (2011)
  • T. Singer et al.

    A common role of insula in feelings, empathy and uncertainty

    Trends Cogn. Sci.

    (2009)
  • J. Wagemans et al.

    Introduction to Michotte's heritage in perception and cognition research

    Acta Psychol.

    (2006)
  • M. Wiener et al.

    The image of time: a voxel-wise meta-analysis

    NeuroImage

    (2010)
  • D.H. Wu et al.

    Neural substrates of processing path and manner information of a moving event

    Neuropsychologia

    (2008)
  • D.M. Amodio et al.

    Meeting of minds: the medial frontal cortex and social cognition

    Nat. Rev. Neurosci.

    (2006)
  • S.J. Blakemore et al.

    How the brain perceives causality: an event-related fMRI study

    Neuroreport

    (2001)
  • S.J. Blakemore et al.

    The detection of contingency and animacy from simple animations in the human brain

    Cereb. Cortex

    (2003)
  • L. Brothers

    The social brain: a project for integrating primate behavior and neurophysiology in a new domain

    Concepts Neurosci.

    (1990)
  • M. Brune

    "Theory of mind" in schizophrenia: a review of the literature

    Schizophr. Bull.

    (2005)
  • M. Carrozzo et al.

    Tempo rubato : animacy speeds up time in the brain

    PLoS One

    (2010)
  • Cited by (15)

    • Intuitive physics ability in systemizers relies on differential use of the internalizing system and long-term spatial representations

      2018, Neuropsychologia
      Citation Excerpt :

      Intuitive physics expertise can also be approached from neurocognitive functions directly involved in physical processing, such as processing of mechanistic movement and making causal inferences about moving physical objects. Performance of tasks involving, for example, predicting the direction toward which a shaky tower will fall are associated with activation of the frontoparietal network, as well as specific temporal cortex areas (Blos et al., 2012; Fischer et al., 2016; Fugelsang et al., 2005; Jack et al., 2013; Wende et al., 2013). However, the frontoparietal network is not specific to intuitive physics whatsoever; instead, this domain-independent network is required in almost any task (e.g., Duncan, 2010), and its activity is proportional to cognitive load (Owen et al., 2005).

    • Semantic Processing

      2015, Brain Mapping: An Encyclopedic Reference
    • Neural basis of altered physical and social causality judgements in schizophrenia

      2015, Schizophrenia Research
      Citation Excerpt :

      Integration dysfunctions during motion perception in psychosis have previously been suggested for both social (abstract) motion attributes (Kim et al., 2005; Kim et al., 2011) and physical parameters, e.g. time (Volz et al., 2001; Papageorgiou et al., 2013). Visual–spatial motion characteristics are processed by the visual system in the occipital lobe (Blakemore et al., 2001; Fonlupt, 2003; Badler et al., 2010) and further integrated by posterior (occipito-parietal) regions (Straube and Chatterjee, 2010; Straube et al., 2011; Blos et al., 2012). Especially in the social judgement task, L.IFG-connectivity directly correlated with patients' causality judgements.

    View all citing articles on Scopus
    View full text