The superior colliculus subserves interhemispheric neural summation in both normals and patients with a total section or agenesis of the corpus callosum

https://doi.org/10.1016/j.neuropsychologia.2004.04.011Get rights and content

Abstract

To verify the possibility that the superior colliculus (SC) subserves interhemispheric neural summation, we presented single or double white visual targets to one or both hemifields in normal participants and in patients lacking the corpus callosum (one with total callosotomy and one with callosal agenesis). Simple reaction time was typically faster with double than single stimuli, a phenomenon known as the redundant target effect (RTE); moreover, confirming previous results, we found a larger RTE in patients without callosum than in normals. In both groups, the redundancy gain was related to neural coactivation rather than to probability summation. The novel finding was that, when using monochromatic purple stimuli that are invisible to the SC, we found a similar redundancy gain in both groups; moreover, this redundancy gain was probabilistic rather than neural. Control experiments with monochromatic red stimuli yielded a RTE of the neural type similar to that with white stimuli and this confirmed that the probabilistic RTE found was specific for purple stimuli. In conclusion, visual input to the SC is necessary for interhemispheric neural summation in both normals and in individuals without the corpus callosum while probabilistic summation can occur without a collicular contribution.

Introduction

Reaction time (RT) to a single stimulus is typically slower than to two or more identical stimuli, a phenomenon known as “redundant target effect” (RTE). Two models have been proposed to explain the RTE, a probabilistic (Raab, 1962) and a coactivation model (Miller, 1982, Miller, 1986). The former postulates that two or more redundant stimuli are processed in separate channels in a horse-race fashion. On each presentation the fastest stimulus wins the race and triggers the response; therefore, on average, multiple stimuli are more likely to yield a fast response than single stimuli. In this model there is no need for redundant stimuli to converge at some stage of processing before a response is made while an important prerequisite is that the various channels are independent. In contrast, the coactivation model postulates a convergence between the redundant stimuli. This convergence ultimately results in a faster response to redundant than to single stimuli. Miller, 1982, Miller, 1986 has developed a mathematical means to decide between the two models by calculating the so-called race-model violation. Briefly, Miller’s race-inequality test, which uses the cumulative frequency distribution (CDF) of RTs, sets an upper limit for probability summation when redundant targets are presented. If this limit is violated, then a probabilistic explanation is no longer tenable and the RTE can be ascribed to a co-activation mechanism that is commonly referred to as a “neural” mechanism. It should be pointed out that violation of probability models is evidence for neural coactivation; non-violation, however, might still be compatible with some moderate amount of neural coactivation not exceeding the upper bound given by the inequality.

The literature on the RTE in normal participants provides examples of redundancy gain that can be explained either by the race model (Corballis, 1998; Corballis, Hamm, Barnett, & Corballis, 2002; Murray, Foxe, Higgins, Javitt, & Schroeder, 2001; Reuter Lorenz, Nozawa, Gazzaniga, & Hughes, 1995; Roser & Corballis, 2002) or by neural coactivation (Cavina-Pratesi, Bricolo, Prior, & Marzi, 2001; Diederich & Colonius, 1987; Forster, Cavina-Pratesi, Aglioti, & Berlucchi, 2002; Iacoboni & Zaidel, 2003; Miller, 1982, Miller, 1986; Miniussi et al., 1998; Savazzi & Marzi, 2002; Schwarz & Ischebeck, 1994). The reason for this discrepancy is unclear and is probably related to specific stimulus and task conditions. As an example, in a recent series of experiments, Turatto, Mazza, Savazzi, and Marzi (2004) have shown that the mechanism underlying the RTE can shift from neural to probabilistic depending upon whether target detection is carried out mainly by the magno- or by the parvocellular visual stream, respectively. In contrast to these different results in normals, there is common agreement that patients with a section or agenesis of the corpus callosum have an enlarged redundancy gain often attributable to neural coactivation (Marzi et al., 1997b, Marzi et al., 1997a; Pollmann & Zaidel, 1999; Reuter Lorenz et al., 1995; Roser & Corballis, 2002). Interestingly, in a study by Iacoboni, Ptito, Weekes, and Zaidel (2000) some callosal patients showed a neural and others a probabilistic summation. In such study violation of probability summation was not associated with a specific type of callosal lesion. However, fMRI evidence was provided that patients with a neural summation showed an activation in extrastriate cortex while this was not the case in those showing probability summation. By the same token in another study, only a patient with a section of the posterior corpus callosum showed an enlarged RTE of a neural type while this was not the case in a patient with an anterior callosal section (Corballis, Corballis, & Fabri, 2003). Furthermore, Barr and Corballis (2003) have provided evidence in a callosal agenesis patient for a role of an enlarged anterior commissure in compensating for the lack of the corpus callosum therefore preventing an abnormally increased RTE.

How can one explain the paradoxically large redundancy gain in callosal patients? One possibility is that the RTE has a different neural locus in normals and in patients without the corpus callosum. In split-brain patients Iacoboni et al. (2000) provided fMRI evidence indicating the SC and the extrastriate cortex as possible sites of summation, while Roser and Corballis (2002) indicated the pons or the reticular formation as more likely sites because of the lack of retinotopy typical of the RTE. One should note, however, that the deep layers of the SC have large receptive fields and therefore they also have a rather coarse retinotopy compatible with the lack of spatial selectivity of the RTE. In normals, Iacoboni and Zaidel (2003) have argued for a premotor locus of the redundancy gain while Savazzi and Marzi (2002) have proposed as candidate the SC, that is, a structure that, among other characteristics, is provided with multimodal neurones, a property that is typical of the RTE (Nickerson, 1973). Additional evidence for a possible visual site comes from a neurophysiological study (Miniussi et al., 1998) showing a selective involvement of extrastriate cortex, an area that is heavily interconnected with the SC. Moreover, Tomaiuolo, Ptito, Marzi, Paus, and Ptito (1997) found an implicit RTE in hemispherectomy patients when one stimulus in a pair was presented to the hemifield corresponding to the ablated hemisphere. Since the SC was intact in these patients this strongly argues for a collicular mediation.

The overall picture emerging from these studies is far from clear and therefore we decided to test directly the hypothesis that the SC mediates neural summation in both normals and in individuals with a total section or agenesis of the corpus callosum. To do so, we took advantage of the neurophysiological notion that the visual pathways originating from short wave sensitive cones (S-cones) do not send or send very few afferents to the SC (de Monasterio, 1978; Marrocco & Li, 1977; Schiller & Malpeli, 1977; Sumner, Adamjee, & Mollon, 2002). The logic underlying our experiment was to use purple monochromatic stimuli that are uniquely detected by S-cones and that are processed by a component of the koniocellular pathway projecting from the lateral geniculate nucleus (LGN) to the cytochrome-oxidase blobs of primary visual cortex (V1) (Hendry & Reid, 2000). The koniocellular pathway makes up a third functional channel in primate LGN and forms three pairs of layers in macaques, one of which (the middle pair), relays input from S-cones to V1 blobs. If the hypothesis that the SC subserves interhemispheric (neural) summation is correct, we should not obtain a neural RTE when using a pair of purple stimuli or a mixed pair of white and purple stimuli since the SC does not receive S-cones input. The first two experiments tested this possibility in normals, the second two experiments in subjects without the corpus callosum, one with a total callosotomy and another with callosal agenesis.

Section snippets

Experiment 1

Normal participants were asked to press a key as quickly as possible following presentation of one white or purple stimulus or of two white, or purple or mixed stimuli. The five conditions of stimulus presentation were randomised. Single stimuli were presented either to the right or to the left visual hemifield; double stimuli were presented simultaneously to both hemifields.

Participants

Eight right-handed (four females) naive students took part in the experiment. Their age ranged between 21 and 33 years and they had normal or corrected-to-normal visual acuity. All gave informed consent and the experiment was carried out according to the principles laid out in the 1964 Declaration of Helsinki.

Apparatus, stimuli and procedure

The participant was seated in front of a PC monitor (Sony Trinitron Multiscan E530) with the eyes at 57 cm from the screen. The initial event in a trial was the presentation of a warning

Results

Mean RTs in the various conditions of stimulus presentation are shown in Fig. 2A.

RT scores were statistically analysed with a two-ways ANOVA with numerosity (single versus double stimuli) and colour (white versus purple stimuli) as factors. The main effect of numerosity, as expected, was significant, F(1, 7) = 240.9, P < 0.001, with double stimuli (376.0 ms) yielding overall faster RTs than single stimuli (398.7 ms). In contrast, there was neither a significant main effect of colour nor a

Experiment 2

In principle, it could be argued that the presence of a neural coactivation when the RTE is tested with white stimuli and a probabilistic summation with purple stimuli might be related to polychromatic stimuli in general and therefore any monochromatic stimulus might yield a probabilistic summation. To check this possibility, we carried out an experiment using red monochromatic stimuli together with white and red-white mixed stimuli.

Participants

Eight right-handed (four females) naive participants different from those of the previous experiment took part in experiment 2. Their age ranged between 20 and 34 years and they had normal or corrected-to-normal visual acuity. All gave informed consent and the experiment was carried out according to the principles laid out in the 1964 Declaration of Helsinki.

Apparatus, stimuli and procedure

Apparatus and procedure were the same as in experiment 1. The stimulus conditions differed only with respect to the use of red stimuli

Results

As in experiment 1, a two-way ANOVA showed a main effect of numerosity, F(1, 7) = 52.6, P < 0.001, with double stimuli (353.8 ms) overall faster than single stimuli (382.9 ms), see Fig. 3A.

The main effect of colour and the first-order interaction numerosity by colour were not significant. As in experiment 1, the redundancy gain was similar in the three conditions: white = 30 ms, red = 27 ms and mixed = 30.2 ms. However, in contrast to experiment 1 we found a violation of the race inequality for all

Experiment 3

The results of the first two experiments in normal participants have suggested that the SC might be a likely locus for the neural coactivation responsible of the interhemispheric summation of visual stimuli. This conclusion was drawn on the basis of a lack of neural coactivation with stimuli that are invisible to the SC. A clear prediction is that if a subcortical centre such as the SC is responsible for interhemispheric neural summation this should also be the case following section or genetic

DDV (total callosotomy)

DDV showed a consistent redundancy gain across all three stimulus conditions: white = 61.1 ms, purple = 20.8 ms, mixed 37.0 ms, see Fig. 4A. Although the redundancy gain with white stimuli was larger than with the other stimuli this difference was not statistically reliable. As expected, there was a significantly (as assessed by one-sample t-tests) larger redundancy gain with respect to the normal participants tested in experiment 1 when considering white stimuli, T(7) = 16.2, P < 0.001 as well as

Experiment 4

Following the same logic as in experiment 2 we tested the possibility that the different result found with purple as opposed to white stimuli might be ascribed to the use of monochromatic stimuli in general. To check that we tested the RTE in the two callosal patients with red rather than purple stimuli. The red stimuli were identical to those used in experiment 2.

DDV

All three colour conditions showed a statistically similar redundancy gain (white = 84.8 ms, red = 67.1 ms, mixed = 46.3 ms), see Fig. 6A, that was reliably larger than that of the normal participants tested in experiment 2: white, T(7) = 14.1, P < 0.001; red, T(7) = 8.9, P < 0.001; mixed, T(7) = 3.4, P < 0.05.

Miller’s disequation yielded a violation of the race-inequality for all three double stimuli, see Fig. 6B. A one-way repeated-measures ANOVA showed that colour was not significant and this

Discussion

The present report provides novel evidence that interhemispheric neural summation is subserved by the SC both in normals and in patients lacking the corpus callosum. When the visual input to the SC is minimised, if not completely abolished, as when short-wavelength monochromatic stimuli are used, neural coactivation no longer occurs in normals and callosal patients. Moreover, with purple stimuli there is no enlarged redundancy gain in callosal patients, see Table 1 summarising the data of the

Acknowledgements

We wish to warmly thank Prof. Mara Fabri for providing invaluable help for the testing of DDV; we are also grateful to Dr. Aldo Paggi for referring to us patient DDV and to Dr. Gabriele Polonara for providing the MRI documentation of the same patient.

References (49)

  • S. Aglioti et al.

    Incomplete gustatory lateralization as shown by analysis of taste discrimination after callosotomy

    Journal of Cognitive Neuroscience

    (2000)
  • M.S. Barr et al.

    Redundancy gain in the acallosal brain

    Neuropsychology

    (2003)
  • N. Berman et al.

    Cortical suppression of ritino-collicular pathway in the monocularly deprived cat

    Journal of Physiology

    (1976)
  • J. Birch et al.

    New method based on random luminance masking for measuring isochromatic zones using resolution colour dislays

    Ophthalmic and Physiological Optics

    (1992)
  • C. Cavina-Pratesi et al.

    Redundancy gain in the stop-signal paradigm: Implications for the locus of coactivation in simple reaction time

    Journal of Experimental Psychology: Human Perception and Performance

    (2001)
  • M.C. Corballis

    Interhemispheric neural summation in the absence of the corpus callosum

    Brain

    (1998)
  • M.C. Corballis et al.

    Paradoxical interhemispheric summation in the split brain

    Journal of Cognitive Neuroscience

    (2002)
  • M.C. Corballis et al.

    Redundancy gain in simple reaction time following partial and complete callosotomy

    Neuropsychologia

    (2003)
  • Z. Daskalakis et al.

    The mechanisms of interhemispheric inhibition in the human motor cortex

    Journal of Physiology (London)

    (2002)
  • F.M. de Monasterio

    Properties of ganglion cells with atypical receptive-field organisation in retina of macaques

    Journal of Neurophysiology

    (1978)
  • R. Desimone et al.

    A role for the corpus callosum in visual area V4 of the macaque

    Visual Neuroscience

    (1993)
  • A. Diederich et al.

    Intersensory facilitation in the motor component? A reaction time analysis

    Psychological Research

    (1987)
  • M. Fabri et al.

    Role of the corpus callosum in the somatosensory activation of the ipsilateral cerebral cortex: An fMRI study of callosotomized patients

    European Journal of Neuroscience

    (1999)
  • A. Ferbert et al.

    Interhemispheric inhibition of the human motor cortex

    Journal of Physiology

    (1992)
  • Cited by (61)

    • Visual interhemispheric communication and callosal connections of the occipital lobes

      2014, Cortex
      Citation Excerpt :

      In the latter case the redundancy gain significantly exceeds that predicted by a simple race model. A redundancy gain compatible with probabilistic summation is typical of observers with intact brains submitted to cross-midline double visual stimulation, whereas in the same situation observers with a surgically split-brain or agenesis of the corpus callosum show an enhanced redundancy gain suggestive of neural summation (e.g., Reuter-Lorenz et al., 1995; Corballis, 1998; Roser and Corballis, 2002, 2003; Iacoboni and Zaidel, 2003; Corballis et al., 2004; Savazzi and Marzi, 2004). One suggested possibility is that an intact callosum inhibits a putative pathway for neural summation, but other possibilities have been considered.

    • Unlike in clinical blindsight patients, unconscious processing of chromatic information depends on early visual cortex in healthy humans

      2014, Brain Stimulation
      Citation Excerpt :

      Similarly, in these trials the participants typically refrained from pressing the RT response button (89% of trials). We used a permutation test [39] to examine whether the RTE can explained by statistical facilitation associated with processing multiple targets, or whether the RTE was caused by neural summation (for details, see Refs. [26,39,40]). The test included 10 001 permutations, and it was focused on early (5–30) percentiles.

    View all citing articles on Scopus
    View full text