Elsevier

NeuroImage

Volume 38, Issue 2, 1 November 2007, Pages 367-385
NeuroImage

Am I safe? The ventrolateral prefrontal cortex ‘detects’ when an unpleasant event does not occur

https://doi.org/10.1016/j.neuroimage.2007.07.044Get rights and content

Abstract

The ventrolateral prefrontal cortex (VLPFC) is implicated in contingency detection and the evaluation of emotionally significant stimuli. However, the mechanisms whereby an individual can effectively avoid painful or unpleasant events are not well understood. We therefore examined whether the absence of an unpleasant somatosensory stimulus could evoke a response in the human VLPFC as a correlate of contingency detection (the feeling that “I am safe”) without any immediately preceding stimulus. In a differential trace-conditioning paradigm, the unpleasant stimulus followed the partially reinforced stimulus in 50% of trials after 3 s; it never occurred after the nonreinforced stimulus. High-resolution DC electroencephalography, current source density mapping, and spatio-temporal source analysis were performed.

After the nonreinforced stimulus, a highly significant negativity over the VLPFC began about a second after the time for the unpleasant stimulus to occur had passed. We concluded that the VLPFC can be activated merely by a sequence of stimuli (with long interstimulus intervals) without any directly preceding stimulus, provided that this sequence creates the expectation that at a certain time an unpleasant stimulus might occur. This mechanism might allow for the detection of conditions under which harmful events could be avoided.

Moreover, in reinforced trials, we found a highly significantly lateralized negativity (N700) that outlasted the strong, unpleasant somatosensory stimulus for about a second. Topography and source analysis pointed to prolonged activation of the somatosensory system. This processing stage preceded activation of the VLPFC. We concluded that N700 might provide important insights into the time course of somatosensory memory traces.

Introduction

We all need to detect and avoid conditions that might harm us. For example, children who cannot feel pain repeatedly injure themselves. But apart from mere nociception, we must be able to detect circumstances that may lead to rewards as well as those that may cause us to be hurt or punished. This is the role of the “contingency detection system.” It is possible, moreover, that a defective contingency detection system may play a role in the pathogenesis of depression (Tremblay et al., 2005). A better understanding of the cerebral mechanisms that enable us to recognize when harmful stimuli do not occur seems essential to the optimization of already existing cognitive strategies of pain control (Apkarian et al., 2005, Seminowicz et al., 2004) and to the elucidation of the cortical mechanisms of learning. The circuitry involved in the perception of pain (which differs importantly from that involved in non-painful somatosensory perceptions) (Davis, 2000) has been well established, and important pioneering work has been done to explain the mechanisms whereby we evaluate stimuli and their salience (Downar et al., 2002, Downar et al., 2003). However, as yet relatively little is known about how we detect the failure of “bad” stimuli to occur.

The ventrolateral prefrontal cortex (VLPFC) plays an important role in contingency reversal learning and the evaluation of (emotionally relevant) stimuli (Morris and Dolan, 2004, Roberts, 2006, Rolls, 2000), as animal model research and functional neuroimaging have shown. However, in order to establish the conditions under which harmful events can be avoided, we need to detect not only when stimuli occur but also when aversive stimulation is absent. Whether the ventrolateral prefrontal cortex also plays an important role in the process of establishing the contingencies when stimuli are absent is not yet understood. Moreover, different functions of the orbitofrontal cortex – such as stimulus evaluation and contingency establishment – are not easily separated (O’Doherty et al., 2003), as changes in stimulus–reward contingencies go along with changes in the affective meaning of a conditioning stimulus: when a conditioning stimulus is suddenly followed by a reward, its affective meaning is also changed.

Thus, in the present paper, we address the question whether the absence of an unpleasant stimulus in a sequence of stimuli with long interstimulus intervals – i.e., contingency evaluation without any preceding stimulus to be evaluated – might suffice to evoke activation of the VLPFC.

An alternative candidate region for the detection of an omitted stimulus in a sequence of stimuli could be the anterior cingulate, which has been shown to be necessary to discriminate between stimuli and to monitor actions and errors (Cardinal et al., 2003, Gehring and Fencsik, 2001).

A postimperative negative variation (PINV) has been described in paired stimulus paradigms (Eikmeier and Lodemann, 1994, Klein et al., 1998, Rockstroh et al., 1999, Wagner et al., 1996) under conditions of loss or gain of control: that is, when, for example, a painful imperative stimulus could suddenly no longer be switched off or, the other way around, suddenly could be (Elbert et al., 1982, Kathmann et al., 1990). A PINV was also described when subjects could not be certain whether their response after the second (imperative) stimulus was correct (Klein et al., 1996a, Klein et al., 1996b). The PINV, which is distributed over anterior scalp regions, has thus been interpreted as a correlate of the detection of changes in contingency (Klein et al., 1998, Rockstroh et al., 1999). We have described two subcomponents of PINV: an early motor PINV maximum over the contralateral central area (N700; motor/proprioceptive post-processing), and a later maximum over ventrolateral prefrontal areas, which might be related to contingency detection (Bender et al., 2004, Bender et al., 2006). Since the VLPFC has been shown to be important in contingency reversal learning, as mentioned above (Morris and Dolan, 2004, Roberts, 2006, Rolls, 2000), an equivalent of PINV over the ventrolateral prefrontal areas could develop after the nonreinforced stimulus in a differential trace conditioning paradigm, indicating that the brain has detected the non-occurrence of the unconditioned stimulus.

We examined such a differential trace conditioning paradigm with an unpleasantly strong somatosensory stimulus following a partially reinforced visual stimulus in 50% of trials after a fixed interval of 3 s, while it never occurred after the nonreinforced visual stimulus. We hypothesized that

  • a late negativity over the ventrolateral prefrontal areas (Bender et al., 2004, Bender et al., 2006) would also occur after the nonreinforced stimulus when the time for the unpleasant stimulus to occur has passed. This ventrolateral prefrontal activation after an “absent” stimulus would indicate that the subject had noted that no unpleasant stimulus had occurred at the crucial time point after the nonreinforced stimulus. The high time-resolution of dense-sensor-array DC EEG would allow imaging of the temporal sequence of cortical activation.

  • contingency awareness (Tabbert et al., 2006) would influence VLPFC activation. We tested for differences between the first block of the recording session, when subjects might still not be aware or fully aware of the correct contingencies and the following blocks when subjects might have already learned the contingencies in our differential trace conditioning paradigm (which stimulus indicated that they were “safe” and which stimulus indicated that they would receive an unpleasant electrical stimulus with a probability of 50%).

Shorter time intervals (short-term integration of somatosensory feedback through sensory memory) might be bridged by different cortical mechanisms. We recently discovered post-processing in secondary motor areas before activation of the prefrontal cortex (motor PINV) (Bender et al., 2004, Bender et al., 2006). This post-processing in secondary cortical areas might occur within the motor system (post-processing of reafferent proprioceptive input), while also representing a general mechanism of brain functioning that is valid for different modalities. A precategorical short-term sensory memory for the visual modality has already been suggested by early psychological investigation (iconic memory) (Sperling, 1963, Sperling, 1967). There have also been discussions about the duration of sensory memory (e.g., Purdy and Olmstead, 1984). So far, no direct neurophysiological correlate of sensory memory has been described apart from indirect hints from mismatch negativity (MMN) in the auditory modality (Deouell et al., 2006, Matuoka et al., 2006). However, one of the basic principles in associative learning is long-term potentiation, which requires two neurons to be simultaneously active in order to establish a connection of increased strength (Hebbian learning) (Cooper, 2005). Short time gaps between subsequent stimuli need to be bridged. This can be done by deliberate attention and working memory. However, many forms of conditioning have been shown to work automatically, especially for short delays. Late modality-specific cortical post-processing could represent the neurophysiological equivalent of short-term sensory memory, which would be supported by our finding of a comparable component (“N700”) after visual stimulation as well (Bender et al., submitted). By analyzing the topography of lateralized potentials, we have been able to eliminate temporal overlaps of widespread endogenous potentials such as P300 and the LPC to a large extent (Bender et al., 2006). We examined possible lateralized activation of primary and secondary somatosensory areas (SI, SII) during the N700 time interval (500 to 1000 ms after stimulus offset according to our previous findings), when the short, rare, strong, unpleasant but not painful somatosensory stimulus had already terminated.

With respect to N700, we hypothesized that

  • A significant lateralized N700 component would appear in the same latency range as for the motor and visual modality (about 500 to 1000 ms after somatosensory stimulation).

  • This lateralized somatosensory N700 would differ in its topography from the visual or motor N700 and would be compatible with an activation of secondary somatosensory areas. Late activation of the secondary cortical areas of the corresponding modality might represent a mechanism of the sensory memory valid for a variety of modalities.

  • No comparable lateralized N700 would be evoked by a visual warning stimulus presented in the center of the visual field.

Section snippets

Subjects

Twenty-seven healthy adult right-handed (according to the Edinburgh Handedness Inventory) (Oldfield, 1971) subjects (15 females, 12 males) between 20 and 58 years of age (mean 37.4 ± 13.4 years) took part in the experiment. None were taking any psychoactive drugs or suffered from neurological or psychiatric disease. All subjects provided informed written consent according to the Declaration of Helsinki. The study was approved by the local ethics committee.

Task and procedure

Two different symbols 5 cm in diameter

Equivalents of PINV over ventrolateral prefrontal areas

A bilateral negativity over ventrolateral prefrontal areas with a left-sided preponderance occurred as equivalents of PINV following the unpleasant somatosensory stimulus as well when there was no unpleasant somatosensory stimulus after S2. For the topography of the average referenced data (isopotential-line maps), see Fig. 1 (1800 ms after the somatosensory stimulus and 3000 ms after its failure to occur). The time course of the equivalents of PINV over the left VLPFC (electrode AF7′;

Discussion

Our most important finding was a highly significant activation of the contralateral VLPFC when the time for the unconditioned stimulus to occur had passed in our partially reinforced differential trace-conditioning paradigm. This VLPFC activation did not seem to depend on the degree of consciousness of the stimulus contingencies throughout the three trial blocks.

Moreover, a highly significant lateralized contralateral negativity about 500 to 1000 ms after the unpleasant somatosensory stimulus

Conclusions

The VLPFC can be activated without any directly preceding stimulus by a sequence of temporally separated stimuli (long time intervals) if this sequence creates the expectation that at a certain time an unpleasant stimulus might occur but does not. The VLPFC thus helps us to recognize conditions in our environment that might lead to unpleasant or painful events and allows us to avoid them.

Non-overlapping subsequent stimuli that are separated by shorter time intervals (up to one second) can be

References (83)

  • N. Fujiwara et al.

    Second somatosensory area (SII) plays a significant role in selective somatosensory attention

    Brain Res. Cogn. Brain Res.

    (2002)
  • C.M. Gomez et al.

    Preparatory visuo-motor cortical network of the contingent negative variation estimated by current density

    NeuroImage

    (2003)
  • C.M. Gomez et al.

    Fronto-parietal networks activation during the contingent negative variation period

    Brain Res. Bull.

    (2007)
  • G. Gratton et al.

    A new method for off-line removal of ocular artifact

    Electroencephalogr. Clin. Neurophysiol.

    (1983)
  • S. Grimm et al.

    Segregated neural representation of distinct emotion dimensions in the prefrontal cortex—an fMRI study

    NeuroImage

    (2006)
  • K. Hoechstetter et al.

    Interaction of tactile input in the human primary and secondary somatosensory cortex—a magnetoencephalographic study

    NeuroImage

    (2001)
  • M.X. Huang et al.

    A parietal-frontal network studied by somatosensory oddball MEG responses, and its cross-modal consistency

    NeuroImage

    (2005)
  • G.D. Iannetti et al.

    Operculoinsular cortex encodes pain intensity at the earliest stages of cortical processing as indicated by amplitude of laser-evoked potentials in humans

    Neuroscience

    (2005)
  • M. Kanda et al.

    Pain-related and cognitive components of somatosensory evoked potentials following CO2 laser stimulation in man

    Electroencephalogr. Clin. Neurophysiol.

    (1996)
  • C. Klein et al.

    Contingent negative variation (CNV) and determinants of the post-imperative negative variation (PINV) in schizophrenic patients and healthy controls

    Schizophr. Res.

    (1996)
  • A. Kok et al.

    Effects of visual-field and matching instruction on event-related potentials and reaction time

    Brain Cogn.

    (1985)
  • V. Kunde et al.

    Topography of middle-latency somatosensory evoked potentials following painful laser stimuli and non-painful electrical stimuli

    Electroencephalogr. Clin. Neurophysiol.

    (1993)
  • F.A. Lenz et al.

    Different generators in human temporal-parasylvian cortex account for subdural laser-evoked potentials, auditory-evoked potentials, and event-related potentials

    Neurosci. Lett.

    (2000)
  • F. Mauguiere et al.

    Activation of a distributed somatosensory cortical network in the human brain: a dipole modelling study of magnetic fields evoked by median nerve stimulation: Part II. Effects of stimulus rate, attention and stimulus detection

    Electroencephalogr. Clin. Neurophysiol.

    (1997)
  • S.D. Mayhew et al.

    Automated single-trial measurement of amplitude and latency of laser-evoked potentials (LEPs) using multiple linear regression

    Clin. Neurophysiol.

    (2006)
  • J.S. Morris et al.

    Dissociable amygdala and orbitofrontal responses during reversal fear conditioning

    NeuroImage

    (2004)
  • R.C. Oldfield

    The assessment and analysis of handedness: the Edinburgh inventory

    Neuropsychologia

    (1971)
  • B. Rahm et al.

    Tracking the subprocesses of decision-based action in the human frontal lobes

    NeuroImage

    (2006)
  • A.C. Roberts

    Primate orbitofrontal cortex and adaptive behaviour

    Trends Cogn. Sci.

    (2006)
  • M. Rugg et al.

    Interhemispheric asymmetries in the visual evoked response: effects of stimulus lateralisation and task

    Biol. Psychol.

    (1978)
  • D.A. Seminowicz et al.

    Cognitive modulation of pain-related brain responses depends on behavioral strategy

    Pain

    (2004)
  • H. Shibasaki

    Central mechanisms of pain perception

    Clin. Neurophysiol., Suppl.

    (2004)
  • G. Sperling

    Successive approximations to a model for short term memory

    Acta Psychol. (Amst)

    (1967)
  • A. Stancak et al.

    Source activity in the human secondary somatosensory cortex depends on the size of corpus callosum

    Brain Res.

    (2002)
  • A. Stancak et al.

    Desynchronization of cortical rhythms following cutaneous stimulation: effects of stimulus repetition and intensity, and of the size of corpus callosum

    Clin. Neurophysiol.

    (2003)
  • A. Stancak et al.

    EEG source analysis and fMRI reveal two electrical sources in the fronto-parietal operculum during subepidermal finger stimulation

    NeuroImage

    (2005)
  • K. Tabbert et al.

    Dissociation of neural responses and skin conductance reactions during fear conditioning with and without awareness of stimulus contingencies

    NeuroImage

    (2006)
  • R.D. Treede et al.

    Late somatosensory evoked cerebral potentials in response to cutaneous heat stimuli

    Electroencephalogr. Clin. Neurophysiol.

    (1988)
  • R.D. Treede et al.

    Cortical representation of pain: functional characterization of nociceptive areas near the lateral sulcus

    Pain

    (2000)
  • M. Wagner et al.

    CNV, PINV and probe-evoked potentials in schizophrenics

    Electroencephalogr. Clin. Neurophysiol.

    (1996)
  • X. Xu et al.

    Pain-related somatosensory evoked potentials following CO2 laser stimulation of foot in man

    Electroencephalogr. Clin. Neurophysiol.

    (1995)
  • Cited by (13)

    • What makes somatosensory short-term memory maintenance effective? An EEG study comparing contralateral delay activity between sighted participants and participants who are blind

      2022, NeuroImage
      Citation Excerpt :

      We used a shorter time interval than previous work (e.g., Vogel and Machizawa, 2004, Katus and Eimer, 2018; Katus et al., 2015) in order to obtain better temporal resolution and to avoid strong overlaps of the analyzed time windows across experimental conditions. Our selection of this specific time window of tCDA was guided by previous work on the sensory post-processing N700 component, which might be related to CDA and other memory-maintenance-related slow waves (Bender et al., 2007; Bender et al., 2010), and by work on early CNV (Bender et al., 2004). On the basis of former research (Postle, 2006; Ruchkin et al., 2003) we investigated neural activity in the frontal and the somatosensory cortex, respectively representing the supramodal 'monitoring’ area and the modality-specific region involved in somatosensory memory maintenance.

    • Stimulus probability affects the visual N700 component of the event-related potential

      2020, Clinical Neurophysiology
      Citation Excerpt :

      The visual occipito-temporal N700 is a negative slow wave with latencies of approximately 550–1100 ms from onset of a short visual stimulus and a peak at approximately 700 ms (Bender et al., 2008; Bender et al., 2010; Hecht et al., 2016). Similar negative slow waves with modality-specific topographic maxima occur after single auditory and somatosensory stimuli as well as after movements (Bender et al., 2007; Bender et al., 2010; Bender et al., 2012a). Already around 1980, it was recognized that slow negative waves followed nonpaired stimuli and contributed to early contingent negative variation (CNV) (Rohrbaugh et al., 1979; Rohrbaugh et al., 1980).

    • Viewing socio-affective stimuli increases connectivity within an extended default mode network

      2017, NeuroImage
      Citation Excerpt :

      The left vlPFC (IFG) showed the highest outward-degree centrality (Fig. 5B and C). The vlPFC has been associated with the expectation and evaluation of emotional stimuli (Hynes et al., 2006; Bender et al., 2007), with decision making (Rolls, 2000; Kringelbach and Rolls, 2004), behavioral inhibition (Rubia et al., 2003; Aron et al., 2004; Goldstein et al., 2007) and emotion regulation (Ohira et al., 2006; Ochsner and Gross, 2008). The left vlPFC appears as a crucial connector hub (high outward degree) which integrates information from a multitude of DMN regions.

    • Daily physical complaints and hippocampal function: An fMRI study of pain modulation by anxiety

      2012, NeuroImage
      Citation Excerpt :

      And the network would be also related to somatosensory amplification or somatization. We identified pain-responsive regions that were correlated with activity in the brain areas implicated in the anxious level of the conditioned cue, because such regions are supposed to include regions associated with cognitive and emotional processing of pain, such as the insula (Baliki et al., 2009; Peltz et al., 2011; Peyron et al., 2000), and the parietal operculum (Benuzzi et al., 2008; Sawamoto et al., 2000; Treede et al., 1999), the anterior cingulate cortex(Critchley, 2004; Medford and Critchley, 2010; Peyron et al., 2000; Singer et al., 2004), the inferior prefrontal gyrus (Bender et al., 2007; Salomons et al., 2007; Wiech et al., 2006), and the middle prefrontal gyrus (Lorenz et al., 2003; Moriguchi et al., 2007; Richter et al., 2010). We supposed that differential neural response to the high and low anxiety cue in the anxiety-related areas was correlated with differential neural response to aversive stimuli with high and low anxiety in pain-related areas.

    View all citing articles on Scopus
    View full text