AWA: Academic Writing at Auckland

The presence of cerebral asymmetry has long been established in the functioning of the human brain. More specifically, left hemispheric dominance in language processing and right dominance in visuospatial (Martin, 1979). This essay examines evolutionary theories postulating reasons for the development of cerebral asymmetries as w ell as anatomical and functional theories regarding its maintenance. There will be a particular focus on Miller and Sergent’s theories though other theories will also be touched on. Whilst extensive studies have been conducted in support of these, the tendency to use right-handed male subjects and the inability to accurately control for the influences of stimulus difficulty may contribute to discrepancies in results.

Interhemispheric transfer is made possible by the corpus callosum, a large fibre tract comprised of axonal fibre bundles connecting the two hemispheres (Loeser & Alvord, 1968). The rate and nature of interhemispheric communication appears to be critical in the development of lateralisation – more specifically, the interhemispheric transfer time (IHTT). This is the rate at which information is relayed between the two hemispheres (Saron & Davidson, 1989) because of axonal conduction velocities (Miller, 1987). A large proportion of corpus callosum fibres have been found to be either finely myelinated or unmyelinated and with average diameters of myelinated axons under 1 μm (Miller, 1987; Ringo, Doty, Demeter, & Simard, 1994).These figures are consistent in studies conducted on cats, rats (Miller, 1987) and macaques (Ringo, Doty, Demeter, & Simard, 1994).

Mean conduction velocities for myelinated axons with diameters of approximately 1 μm and axons with diameters of about 0.5 μm are 0.7m/s and 0.5m/s, respectively. As a result, substantial conduction delays are expected (Miller, 1987). Homotopic interhemispheric distances are approximated to be between 80 – 180mm for direct pathways. It would therefore take an action potential approximately 27msec to cross the hemispheres via a 175mm fibre and even longer for convoluted pathways (Ringo, Doty, Demeter, & Simard, 1994).  As brain size increased along evolutionary lines, distances between homotopic and heterotopic regions increased, subsequently increasing interhemispheric conduction delays (Miller, 1987).

Ringo and colleagues (1994) proposed their temporal economy hypothesis based on these delays. They suggest it is economically and thus evolutionarily advantageous for complex tasks requiring multiple commissural crossings to be specialised in one hemisphere, avoiding unnecessary delays (Ringo, Doty, Demeter, & Simard, 1994). It is an idea founded on efficiency. They say one would not find identical copies of related memories and skills in each hemisphere but rather, unilateral clusters of them. This is because it is unnecessary to have double copies of this information for a task to be carried out, freeing the remaining hemisphere and increasing neural capacity (Ringo, Doty, Demeter, & Simard, 1994). It is important to note that the key experiment in this study was based on a neural network simulation as opposed to human or animal subjects. Though somewhat dynamic in nature, its parameters were set by experimenters and thus may not be an accurate representative of the human brain. In addition to this, it lacked precise measurements for human fibres – a common feature shared by other related studies. Rather, hypotheses are founded on the extrapolation of data gathered from animal models (Miller, 1987; Ringo, Doty, Demeter, & Simard, 1994), calling into question the accuracy to which these findings are able to be applied to human subjects.

The need to multi-task has also been put forward as another possible reason for the development of cerebral asymmetry (Vallortigara & Rogers, 2005). Evidence in support of this comes primarily from animal models, disproving the belief that laterality exists only in humans. For example, both chicks and fish exhibit the importance of being able to use one eye for feeding whilst simultaneously detecting predators with the other. Chicks raised in the light were more lateralised than those raised in the dark. Light-reared chicks were able to perform the aforementioned tasks whilst their counterparts were unable to, leaving them more vulnerable to predators. The same principle applies to fish but it was also found that female lateralised fish were more effective in mating – another evolutionary advantage for survival. However, it was noted that should this behaviour become predictable to predators, the advantage it provides becomes increasingly ineffective (Vallortigara & Rogers, 2005). Again, the extrapolation of these principles to humans seems plausible but it would be necessary for studies focusing on this to be conducted, as opposed to reliance on animal models.

Several hypotheses have also been put forward for both functional and structural substrates underlying laterality. An example of a structural hypothesis is the physiologically-based concept of homotopic callosal inhibition. It suggests that contralateral patterns exist as mirror-images and “photographic negatives” in each hemisphere (Cook, 1984). It is put forward that mirror-imaging is made possible by symmetry in the hemispheres and homotopic callosal connections. Using the example of language, Cook (1984) suggests that excitation in left hemispheric areas causes a suppression of language-related neurons in the right hemisphere where “contextual” or semantic neurons are also excited (Cook, 1984). Whilst this study appears to have the necessary evidence to support it, it important to point out that it is a study focused purely on language. It also assumes that the cerebral hemispheres are generally symmetrical (Cook, 1984) – something which recent DTI studies and Miller (1987) suggest otherwise.

Another structural substrate hypothesis is that of transcortical cell assemblies. Here, it is believed that large populations of highly-connected neurons form functional units and it is here that higher cognitive processes such as concepts and words are stored (Pulvermuller & Mohr, 1996). Some of these clusters may include neurons from both hemispheres; forming transcortical assemblies and others are spread throughout the cortical area. Since assemblies are densely connected via strong intra-assembly connections, activation should register as high-frequency cortical responses on an EEG (Pulvermuller & Mohr, 1996). It is the balance of these assembly neurons between the two hemispheres rather than their location in either hemisphere which has been identified as important. Symmetrical transcortical assembles are present. However, more complex tasks are hypothesised to be transcortical – spread over both hemispheres, and showing varying degrees of assembly laterality. For example, function words such as conjunctions and pronouns would be more strongly lateralised than content words such as concrete nouns (Pulvermuller & Mohr, 1996). Evidence from finger-tapping exercises produced similar results with simple, repetitive movements lateralised and more complex coordination activating bilateral networks. It must be noted, however, that the existence of cell assemblies has not been clearly proven (Pulvermuller & Mohr, 1996) and as with the callosal inhibition study, this study relies heavily on linguistic evidence.   

Hormones, more specifically gonadal steroids in women, have also been shown to influence functional cerebral asymmetries (FCAs). These FCAs have been found to be stable in men but cyclically fluctuating in women. It is hypothesised that these changes occur via interhemispheric inhibition (Weis, Hausmann, Stoffers, Vohn, Kellermann, & Sturm, 2008).  Word-matching tasks and fMRI were used. Women were found to have decreased inhibition and thus FCAs due to increased estradiol levels during the follicular phase of the menstrual cycle. The opposite was true during the menstrual phase where the inhibitory effect of left hemispheric language areas on homotopic areas was strongest (Weis et al. 2008).

Robert Miller’s axonal conduction time theory is prominent in the explanation of both structural and functional substrates underlying laterality. Miller hypothesises that the cortico-cortical fibres in the right hemisphere are more greatly myelinated, have larger diameters and are faster-conducting than those in the left (Iwabuchi & Kirk, 2009; Miller, 1987). That is, the left hemisphere contains a greater repertoire of slow-conducting axons (Barnett, Corballis, & Kirk, 2005). Overall then, the mean time taken for projections to travel from the right to left hemisphere should be less than those travelling in the opposite direction (Miller, 1987). It is this, he says, which underlies the verbal/visuospatial dichotomy via serial and parallel processing – as will be discussed later (Barnett, Corballis, & Kirk, 2005).

EEG work in which interareal coherence and phase lag were compared provides indirect evidence supporting this claim (Thatcher, Krause, & Hrybyk, 1986). In their 1986 paper, Thatcher and colleagues recorded EEGS from various areas of both hemispheres and computed and compared values for the two aforementioned values. They found relatively stronger interareal phase lag in the left hemisphere and greater interareal coherence in the right. While Thatcher and colleagues attributed these findings to longer, more myelinated connections existing in the right (Thatcher, Krause, & Hrybyk, 1986), Miller put forth that the relatively less myelination in the left hemisphere was a result of its having small-diameter axons (Miller, 1987).  The right hemisphere would, by inference, have quicker and more temporally concentrated interareal conduction (Miller, 1987).

The postulated range of axon diameters and lengths mean cortical connections have very different conduction velocities and thus delays. These differences could account for left hemispheric dominance in phoneme discrimination with the encoding of temporal patterns from the primary auditory cortex via a series of anatomical “delay lines” (Miller, 1987). Here, neurons are suggested to receive a range of inputs over a brief period of time. Connections are processed sequentially, beginning with inputs from faster-conducting axons. On the contrary, right hemispheric neurons receive inputs from axons with more uniform velocities which are processed simultaneously (Miller, 1987).  It is on this basis that Miller attributes dominance to the left hemisphere in serial processing and parallel processing to the right (Barnett, Corballis, & Kirk, 2005; Iwabuchi & Kirk, 2009). These are the abilities to process elements sequentially, one-at-a-time and to process elements simultaneously, respectively (Polich, 1982). It could also be called analytical and holistic processing and could account for left hemispheric dominance in language processing and right in visuospatial processing, respectively.

Nevertheless, White and White (1975) suggest that the above differences are limited to tasks involving verbal or visuospatial matching of alphanumeric stimuli or words. Additionally, the left hemisphere, appears to also be capable of parallel processing when a task calls for the analysis of stimulus features (Polich, 1982) – a finding which supports Sergent’s theory. However, it must be noted that all of Polich’s participants were right-handed males. Other studies have yielded interesting results both in support and as an extension of Miller’s hypothesis.

For example, the hypothesis does not seem to apply to schizophrenic patients who may have dysfunctional interhemispheric transfer times (IHTT). A 2005 EEG study showed that schizophrenic patients lacked the usual asymmetrical speed in right-to-left IHTT.  Here, the transfer symmetry observed is attributed to a lack of right hemispheric functional activation, affecting only right-to-left transfers (Barnett, Corballis, & Kirk, 2005). It is possible that this is a reflection of a loss of white matter in the right hemisphere - suggested by Miller (1987) to be a key reason behind faster conductions here. Affected regions include the right interior capsule, right anterior commissure and right hemisphere frontal tracts. However, as noted in the study, all the subjects involved were males and medication was not controlled for (Barnett, Corballis, & Kirk, 2005).

Miller’s hypothesis also does not take into account handedness where there exists a difference in directional asymmetry. Iwabuchi and Kirk investigated interhemispheric transfer time (IHTT) and its directional asymmetry in left-handed individuals. 15 right-handed and 10 left-handed males participated and stimuli presented to either the left (right hemisphere) or right visual field (left hemisphere) whilst EEG activity was recorded. Their results reaffirmed earlier findings and Miller’s hypothesis in right-handers but the same directional asymmetry was not found in left-handed males (Iwabuchi & Kirk, 2009). Left-handers were found to show similar patterns of symmetry to females. That is, with a tendency for bilateral hemispheric activation (Iwabuchi & Kirk, 2009). Once again though, females were not tested and as pointed out in Marzi’s review (2010) and distinctions were not made between the differential writing posturues of left-handers – that is, normal or inverted (Marzi, 2010).

Another prominent functional theory is Sergent’s spatial frequency hypothesis. It is suggested that the left hemisphere better processes later-available local (high spatial frequency) stimuli whilst the right more effectively processes early-available global (lower spatial frequency) stimuli (Sergent, 1982). Support for this includes neuropsychological studies and facial processing experiments. It must be remembered that the layout of the human visual system is such that information processed from one side of the visual field travels to the contralateral hemisphere of the brain (Hellige & Sergent, 1986).

Sergent critiques previous work carried out on the verbal/visuospatial dichotomy. More specifically, work in support of callosal inhibition and those in which visual stimuli were presented for brief periods of time only. She implies that both left and right hemispheres are capable of processing both verbal and visuospatial information (Sergent, 1982).The differences, rather, lies in the amount of sensorimotor resolution needed for processing and the sensorimotor resolution capacities of each respective hemisphere. That is, it is dependent on the quality of the visual stimulus. For example, the left hemisphere dominates when clear stimuli are presented for longer-than-usual durations (Sergent, 1982). One must keep in mind that contained in her critiques are comparisons to previous studies she had carried out on the topic, opening room for a potential bias.

Neuropsychological studies in support of Sergent involved the use of Navon figures where larger letters (global) comprised of smaller letters (local) were presented to subjects (Navon, 1977). Low frequencies were assumed to be detected earlier than high. It was also assumed that if a low frequency stimulus was presented for a long time, it would still be preferentially processed by the right hemisphere (Sergent, 1982). That is, subjects should show a decreased ability to identify the smaller letters when presented to the left visual field (right hemisphere). Physical Identity letter comparison tasks have been used to investigate this. Subjects were presented with two letters and were to decide if they were identical or different. Stimuli were blurred using any of the following techniques: decreasing stimulus luminance, decreasing the length of presentation, increasing physical complexity and presentation of stimuli nearer the periphery (Sergent, 1982). All these acted to remove high spatial frequencies. Results show that left visual field (right hemisphere) performance was unaffected and the contrary was so for the right visual field (left hemisphere) (Whitman & Keegan, 1991).

It was noted that earlier experiments of this nature did not accurately account for the effects remaining from untouched factors. For example, high frequency information may not actually be removed when stimulus duration is decreased (Jonsson & Hellige, 1986). Jonsson and Hellige (1986) mention that the effects of stimuli difficulty and the differentiation between response accuracy and response time effects need to be accounted for (Jonsson & Hellige, 1986). Despite failing to obtain significant results in support of their predicted Stimulus Clarity x Visual Field effect, Jonsson and Hellige do not aim to disprove Sergent’s hypothesis. Rather, to suggest that further study be conducted to tests the parameters of its applicability. On the other hand, Peterzell (1991) more strongly opposes Sergent’s findings, saying there is probably no relation between spatial frequency and hemispheric competence citing. He bases this argument on similar findings to that of Jonsson and Hellige as well as the possibility that hemispheric response bias, rather than specialisation, is responsible for Sergent’s findings (Peterzell, 1991).

Facial processing experiments were carried out by presenting faces to the right or left visual field in a tachistoscopical manner with filtering of high or low spatial frequency components (Whitman & Keegan, 1991). Subjects are asked to discriminate faces which differed from each other on a few features and reaction times for both the same and different conditions were recorded (Sergent, 1982). A positive thing to note about Sergent’s experiment here is her use of both male and female subjects. Nevertheless, the use of line drawings instead of photographs matched for size and features may suggest that the level of applicability is compromised. This is because other cues individuals may use in facial processing are not available – for example, colour. However, it is difficult to guess the level of impact colour would have in a test situation like this. Results generally showed a left visual field (right hemisphere) advantage though this could be dependent on whether or not it was possible for stimuli to be verbalised (Sergent, 1982). Again, Sergent concludes that both hemispheres are capable of processing visuospatial information but differences lie in the characteristics of stimuli. For example, the right hemisphere appears to be influenced by the saliency of features (Sergent, 1982).

It seems plausible to suggest that Miller’s serial/parallel processing and Sergent’s spatial frequency theories are not mutually exclusive and that in reality, are supported by the same circuitry. The term “high frequency” could easily be interchanged with “analytical” or “serial” processing where there is preference for local, detailed components of stimuli. On the other hand “low frequency” could just as easily describe the “holistic” or “parallel” processing of more global stimuli characteristics. Along with that, it does not seem far-fetched to suggest that the anatomical “delay line” machinery described by Miller is able to support these functions.

In terms of reliability, it has been noted that in the majority of studies mentioned in this essay, subjects involved in experiments were predominantly right-handed males (e.g. Jonsson & Hellige, 1986; Miller, 1987; Whitman & Keegan, 1991). It is well-established that asymmetry manifests itself differentially depending on both handedness and sex ( Iwabuchi & Kirk, 2009; Weis, et al. 2008). Females have tended to display faster; more symmetric patterns than males (Nowicka & Fersten, 2001). In addition to this, studies have noted that varying the difficulty of a target stimulus is able to modify results collected (Hellige & Sergent, 1986; Jonsson & Hellige, 1986; Peterzell, 1991).

Works Cited

Barnett, K., Corballis, M., & Kirk, I. (2005). Symmetry of callosal information transfer in schizophrenia: a preliminary study. Schizophrenia Research , 74 (2-3), 171-178.

Cook, N. (1984). Callosal Inhibition: The Key to the Brain Code. Behavioural Science , 29, 98-110.

Hellige, J., & Sergent, J. (1986). Role of Task Factors in Visual Field Asymmetries. Brain and Cognition , 5, 200-222.

Iwabuchi, S., & Kirk, I. (2009). Atypical interhemispheric communication in left-handed individuals. BEhaviour , 20, 166-169.

Jonsson, J., & Hellige, J. (1986). Lateralized effects of blurring: A test of the visual spatial frequency model of cerebral hemisphere asymmetry. Neuropsychologia , 24 (3), 351-362.

Loeser, J., & Alvord, E. (1968). Agenesis of the corpus callosum. Brain , 91 (3), 553-570.

Martin, M. (1979). Hemispheric specialisation for local and global processing. Neuropsychologia , 17, 33-40.

Marzi, C. (2010). Asymmetry of interhemispheric communication. John Wiley & Sons, Ltd. , 1, 433-438.

Miller, R. (1987). Representation of breif temporal patterns, Hebbian synapses, and the left-hemisphere dominance for phoneme recognition. Psychobiology , 15 (3), 241-247.

Navon, D. (1977). Forest Before Trees: The Precedence of Global Features in Visual Perception. Cognitive Psychology , 9, 353-383.

Nowicka, A., & Fersten, E. (2001). Sex-related differencesi n interhemispheric transmission time in the human brain. Neuroreport , 12, 4171-4175.

Peterzell, D. (1991). On the Nonrelation between Spatial Frequency and Cerebral Hemispheric Competence. Brain and Cognition , 15, 62-68.

Polich, J. (1982). Hemispheric differences for visual search: Serial vs parallel processing revisited. Neuropsychologia , 20 (3), 297-307.

Pulvermuller, F., & Mohr, B. (1996). The Concept of Transcortical Cell Assemblies: a Key to the Understanding of Cortical Lateralization and Interhemispheric Interaction. Neuroscience & Biobehavioral Reviews , 20 (4), 557-566.

Ringo, J., Doty, R., Demeter, S., & Simard, P. (1994). Time Is of the Essence: A Conjecture that Hemispheric Specialization Arises from Interhemispheric Conduction Delay. Cerebral Cortex , 4, 331-343.

Saron, C., & Davidson, R. (1989). Visual Evoked Potential Measures of Interhemispheric Transfer Time in Humans. Behavioural Neuroscience , 103 (5), 1115-1138.

Sergent, J. (1982). About Face: Left-Hemispheric Involvement in Processing Physiognomies. Journal of Experimental Psychology: Human Perception and Performance , 8 (1), 1-14.

Sergent, J. (1982). The Cerebral Balance of Power: Confrontation or Cooperation? Journal of Experimental Psychology , 8 (2), 253-272.

Thatcher, R., Krause, P., & Hrybyk, M. (1986). Cortico-cortical associations and EEG coherence - A two-compartment model. Electroencephalography & Clinical Neurophysiology , 64, 123-143.

Vallortigara, G., & Rogers, L. (2005). Survival with an Asymmetrical Brain: Advantages and Disadvantages of Cerebral Lateralization. Behavioural and Brain Sciences , 28, 575-633.

Weis, S., Hausmann, M., Stoffers, B., Vohn, R., Kellermann, T., & Sturm, W. (2008). Estradiol Modulates Functional Brain Organization during the Menstrual Cycle: An Analysis of Interhemispheric Inhibition. The Journal of Neuroscience , 28 (50), 13401-13410.

White, M., & White, G. (1975). Parallel-serial processing and hemispheric function. Neuropsychologia , 13, 377-381.

Whitman, R., & Keegan, J. (1991). Lateralization of Spacial Processing: A Spatial Frequency Model. Intern. J. Neuroscience , 60, 177-185.