Possible Subtypes of Attention Deficits and their Overlap with Specific Learning Disabilities and Behavior Disorders.
A number of richly interconnected anatomic areas in the cerebral cortex and deeper brain structures compose the brain's attentional system. Some of the brain areas involved in the attentional matrix are anatomically near to or congruent with areas having to do with high order information processing. The complexity of the attentional matrix and its anatomic intimacy with information processing raises two basic issues for consideration. The first is that ADHD is probably a collection of several different disorders of attention, each with its own underlying neural mechanism. Second, some of these potential subtypes of ADHD will probably turn out to be characteristically linked with certain specific learning disabilities. Subtypes of ADHD and their possible linking with specific learning disabilities may have important influences on the practice of neurofeedback.
Three Aspects of Attention.
One conceptualization of attention involves three components. First, there is a "state" function; that is, the degree of arousal or attentional tone (Mesulam, 1985-b, pp. 126-7). In severe cases, dysfunction in the state function of attention manifests as a global confusional state or even coma. Less severe cases may manifest as impersistence to task, and as a generally poor mental performance.
Second, there is the orienting response, which functions to search the environment and choose a particular stimulus as a target for attention (Halperin, McKay, Matier, and Sharma, in press). The orienting response is a highly efficient state, in which the brain turns down its energy output to physiologic functions such as blood pressure and heart rate (Lynch, 1985, pp. 155-62) while producing high amounts of cortical beta activity (Andreassi, 1989, p. 217).
Third, there is the "vector" function; that is, directed, selective attention (Mesulam, 1985-b, p. 127). Dysfunction in the vector function of attention manifests as distractibility or an inability to respond appropriately to a stimulus to which one has oriented. However, since attention is always an interaction between these three components, determining which component is malfunctioning, let alone its anatomical correlates, is often all but impossible (Mesulam, 1985-b, pp.126-127).
Anatomy of the Brain's Attention Matrix.
A complex matrix of neural structures caries out the brain's attentional functions. The components of this matrix are all richly interconnected. Each component's function within the matrix is unique to a certain degree. Yet malfunction in one component often produces effects which are similar enough to those produced by malfunction in another component that it is extremely difficult to tell with certainty in which component the malfunction lies (Mesulam, 1985-b, pp.134-140). This is one reason why the exact neurophysiologic causes of ADHD have yet to be determined, for malfunction in one or more anatomic sites may cause what to date has been described as one behavioral syndrome. In all probability, each possible anatomic site is at fault in at least some cases of ADHD. Some sites, such as the right frontal cortex or the reticular activating system, may be more commonly involved. Also, two or more sites may be involved simultaneously in some cases of ADHD.
The reticular activating system (RAS) is the brain's arousal system, regulating wakefulness and sleep. It sets attentional tone and capacity, making it the most important component in the state function of attention. However, it also enhances neural signal-to-noise ratios, an important aspect of directed attention (Mesulam, 1985-b, 135-137). Malfunction in the RAS causes some degree of a confusional state; this can include rather specific deficits in cortical functioning, such as anomia, dysgraphia, dyscalculia, and constructional deficits (Mesulam, 1985-b, p.129). More severe malfunction of the reticular activating system manifests as coma. Everyone has experienced inadequate functioning of their reticular activating system when they are not fully awake, either upon first waking up, or when overly fatigued and in need of sleep. Such may be the experience much of the time for many people with ADHD.
Much of what is called attention deficit-hyperactivity disorder could be explained by the presence of an inadequate, or "underpowered," reticular activating system. A number of lines of evidence are consistent with such a theory. First is the oft-cited observation that many ADHD children seem to be underaroused and in need of stimulation (Satterfield, 1972; Zentall, 1975). Second is the observation, more exactly quantified by Lubar and associates over the past 16 years, that the EEG pattern of ADHD children shows excessive theta activity relative to beta activity (Satterfield, 1972; Mann, Lubar, Zimmerman, Miller, and Muenchen, 1992). Theta activity normally tends to predominate in lighter phases of sleep; thus, the characteristic EEG pattern of ADHD is suggestive of an underaroused state (Satterfield, 1972). Third, the PET scan data of Zametkin and associates (1990) demonstrates a lower metabolic rate in the cerebral cortices of hyperactive adults when compared to non-hyperactive controls; such a hypometabolic cerebral cortex is consistent with underarousal, as one would expect in the case of a weak RAS. And finally, stimulant medications used in the treatment of ADHD, such as methylphenidate (Ritalin) and dextroamphetamine, have a strong excitatory effect on norepinephrine receptors; norepinephrine is the predominant neurotransmitter in the locus cerulus, a part of the RAS having to do with arousal and augmenting neural signal-to-noise ratios (Noback, Strominger, and Demarest, 1991, pp. 241-244; Ziegler and Lake, 1984, pp 110-111).
The prefrontal cortex, especially on the right side, has a number of functions which relate to ADHD. It has an important role in the motor aspects of attention, including orientation to or search for a stimulus, exploration of the environment (Mesulam, 1985-b, p. 157), and impulse inhibition (Stuss and Benson, 1986, p. 243). Also, the prefrontal lobes organize information in sequences, establish sets of information, and integrate new and old information to form novel interpretations (Stuss and Benson, 1986, pp. 242-243); ADHD children typically have great difficulty in these areas of planning and organization. Furthermore, the prefrontal cortex prepares the sensory areas of the cortex to expect incoming data (Comings, 1990, pp. 350-351), an important component of the vector aspect of attention. And in ways not well understood, the right prefrontal lobe also plays a role in attentional tone (Mesulam, 1985-b, p. 140), though its contribution to directed attention is its predominant role in the attentional matrix. Patients with prefrontal lesions, depending on the site of the lesion, show one or more of the following signs and symptoms: impulsiveness, lethargy, poor organization, lack of directedness in motor activity, hyperactivity, hypoactivity, distractibility, impersistence, behavior which is overly dependent on environmental stimulation and lacks inner initiative, confusional states, and a tendency to neglect the left hemispace (Comings, 1990, pp. 341-361; Mesulam, 1985-b, p. 140; Stuss and Benson, p. 224).
An additional attentional function of the prefrontal lobes is that they help regulate the striatum (Comings, 1990, pp. 354-355 and 393-395) and the limbic system (Comings, 1990, p. 344). The striatum is a part of the brain's attentional system which can play a major role in hyperactivity, while the limbic system helps regulate both activity level and emotional control and supplies the motivation to attend (see below). Besides regulation of the limbic system, another emotional role of the right frontal cortex is in the emotional expressiveness of language, or prosody (Ross, 1985, p.247-248).
Thus, the known behavioral functions of the prefrontal cortex suggest that this region might well produce part or all of the symptom complex of ADHD. There are other pieces of evidence which also suggest that this region plays a role in ADHD. First, Hynd and associates demonstrated that hyperactive children have a relatively narrow right prefrontal cortex when compared to non-ADHD controls (Hynd, Serud-Clikeman, Lorys, Novey and Eliopulos, 1990). (Interestingly, in this same study, children diagnosed as "dyslexia pure" had a relatively narrower right prefrontal cortex than did hyperactive children.) Second, the PET scan study by Zametkin and associates (1990) of cerebral metabolism showed that the areas of lowest metabolic activity in hyperactive adults were the prefrontal and premotor areas. Third, in addition to their norepinephrine activity, stimulant medications have powerful dopamine excitatory effects. Dopamine is known to be an important neurotransmitter in the orbitofrontal cortex, an area of the prefrontal cortex which plays an important role in the inhibition of impulsiveness and the regulation of motor activity (Comings, 1990, pp. 389-398; Noback, Strominger, and Demarest, 1991, p. 241; Stuss and Benson, 1986, p. 224). Fourth, patients with certain frontal lobe lesions exhibit behaviors similar to those of children with ADHD and with conduct disorder, which overlaps clinically with ADHD (Comings, 1990, pp. 341-361; Stuss and Benson, 1986, p. 134). Fifth, the EEG evoked potential pattern which is characteristic of the orienting response has a frontocentral location (Knight, 1985, p. 337) and tends to disappear, along with the orienting response, in some patients with prefrontal lesions (Knight, 1985, p. 339). And finally, anecdotal reports indicate that ADHD children who are hyperactive and impulsive tend to have an excess of right frontal theta activity.
The prefrontal cortex also seems to play a role in learning disabilities. An example is dyslexia, as evidenced by the quantitative EEG data of Duffy and associates (1980) and by the MRI study of Hynd and associates cited above. Furthermore, the heteromodal association areas of the prefrontal cortex process the behavioral aspects of sensory information, integrate sensory input with motor output, participate in stimulus discrimination, and handle the planning of complex behaviors (Mesulam, 1985-a, pp. 28-29); thus, prefrontal malfunction might lead to a variety of information processing and specific learning disabilities. In laboratory studies, rats with experimentally induced frontal lobe dopamine deficiency have a decreased ability to learn maze tasks (Comings, 1990, pp. 390-391). However, perhaps because of the complexity of the prefrontal heteromodal association areas, their malfunction is notoriously difficult to describe and quantify (Mesulam, 1985-a, p. 29).
The parietal heteromodal association cortex, especially on the right side, also plays an important role in the attention system. The right parietal cortex contains a bilateral map of sensory space, both intrapersonal and extrapersonal; it also synthesizes sensory input from all modalities into complex concepts (morphogenesis) (Mesulam, 1985-b, pp. 150-152). The parietal heteromodal association cortex also affects attentional tone (Mesulam, 1985, p. 160). Malfunction of the parietal heteromodal association area can cause day dreaming, difficulty in assembling sensory concepts (dysmorphogenesis/amorphogenesis), confusional states, and a tendency to neglect the left hemispace (Mesulam, 1985-b, pp. 131, 150-152, and 160). EEG data also implicates left parietal malfunction in certain types of dyslexia (Duffy,, Denkla, Bartels, and Sandini, 1980). Thus, this brain area also has great potential for causing both attention deficits and learning disabilities.
Regarding a possible parietal location for attention deficits, one additional piece of evidence comes from recent quantitative EEG studies of non-hyperactive ADHD boys (Mann, Lubar, Zimmerman, Miller, and Muenchen, 1992). In the study group, the right posterior surface EEG activity, roughly equivalent to the parietal area, showed increased theta activity when compared to the control group's EEG data. This data is consistent with anecdotal observations that inattentive, non-hyperactive children with ADHD tend to have excessive slow wave activity over the parietal area on quantitative EEG.
The limbic system, especially the cingulate gyrus, communicates the motivational relevance of a stimulus to the cerebral cortex and thus provides the motivation to attend (Mesulam, 1985-b, pp. 153 and 156-158). Lesions in the right cingulate cortex in humans cause neglect of the left hemispace (Mesulam, 1985-b, p. 153) and interfere with temporal ordering or sequencing (Comings, 1990, p. 328). Also, the limbic system is the likely generator of the P300 component of evoked potentials, which is that portion of evoked potentials that is associated with attention (Knight, 1985, pp. 336-338). Furthermore, the limbic system has direct input to the motor system via the nucleus accumbens; lesions of the nucleus accumbens or of the mid-brain ventral tegmental area, which regulates it, can produce disorders of activity level ranging from hypoactivity to hyperactivity (Comings, 1990, pp. 382-383).
Thus, malfunction of the limbic system might produce an attention deficit characterized by lack of motivation to attend, or by assigning inappropriate motivational valence to stimuli (e.g., conduct disorder), as well as by hyperactivity or hypoactivity.
The thalamus provides communication between various components of the attention system and is the switching station for selective attention (Mesulam, 1985-a, pp. 43-46, and 1985-b, p. 157). Theoretically, thalamic malfunction could mimic any of the above disturbances, depending on what area of the thalamus malfunctions. Similarly, lesions in pathways connecting the thalamus with the cerebral cortex or limbic system would cause malfunction in the affected components of the attention system. Furthermore, the reticular nucleus of the thalamus (not to be confused with the reticular activating system) serves to selectively inhibit all thalamic relays other than the one(s) currently chosen by the cortex, especially the prefrontal lobes. Thus the thalamus plays an important role in selective attention which is superimposed on its function as a relay station. This selective inhibition can be overridden by the reticular activating system (Mesulam, 1985-b, p. 138).
The striatum, part of the system of basal ganglia in the forebrain, has a definite, though poorly understood, role in the attention system. It connects with many parts of the thalamus, and thus to the cortex and limbic system, as well as to several subcerebral nuclei (Carpenter and Sutin, 1983, pp. 587-598; Mesulam, 1985, pp. 40-42). It is involved in the organization of complex of motor activities by integrating sensory, emotional, and cognitive inputs into the motor system (Comings, 1990, p. 366); malfunction may result in severe movement disorders, hyperactivity, hypoactivity, deficits in passive avoidance learning, and/or neglect for the contralateral hemispace (Mesulam, 1985-b, pp.156-158; Carpenter and Sutin, 1983, pp. 604-609). Lou and associates (1989) documented striatal hypoperfusion on PET scans in a small group of hyperactive children, several of whom had a history of medical complications at birth. Interestingly, this hypoperfusion was corrected after the patients took stimulant medication. This finding suggests that the cause of the hypometabolism may not have been in the striatum itself, but instead may represent poor functioning in one of the brain structures which regulate the striatum, such as the reticular activating system, the nucleus accumbens, or the prefrontal cerebral cortex.
Besides the above brain areas, any of which can produce some or all of the symptoms of a generalized attention disorder, other areas of the brain affect attention in a more limited scope. The unimodal association areas, each of which processes information from just one sensory modality, are essential components of attention for their respective sensory modality. For example, a lesion in the visual association cortex not only interferes with visual understanding but also causes visual distractibility (Mesulam, 1985-b, pp. 138-139). Thus, unimodal association areas of the cerebral cortex contribute to vector attention within the limited scope of each's sensory modality.
Brief Overview of Neurotransmitters and Drug Therapy
In general, the neurochemistry of attention deficits and related disorders is very complex. To date, no one has discovered a single drug which corrects ADHD precisely and completely, though a certain drug may be remarkably effective in certain cases. This lack of a "silver bullet" for ADHD probably reflects a primarily anatomic basis for attention deficits, rather than a neurochemical basis, at least in most cases. Given the complexity of neural mechanisms, invariably a number of neurotransmitters regulate any one anatomic area, and any one neurotransmitter generally is active in a number of areas of the brain and caries out several functions which often compete with each other. Thus, neurochemical manipulations through the use of one or more drugs may compensate for an anatomic deficit by causing increased activity in an under-functioning area. In most cases, such compensation will not be complete, and probably will occur at the expense of unwanted side-effects. However, as in other brain mechanisms relevant to ADHD, present knowledge of neurotransmitter systems involved in ADHD is incomplete. Future discoveries may provide more evidence of a purely chemical defect, at least in certain subtypes of attention deficits.
Three neurotransmitters seem to be most important in attentional processes, namely, dopamine, norepinephrine, and serotonin. To date, dopamine has received the most attention in medical literature, mainly because stimulant medications such as methylphenidate (Ritalin) have strong stimulating effects on dopamine receptors. Also, rich concentrations of dopamine-responsive neurons are present in the prefrontal cortex, the nucleus accumbens, and the striatum (Comings, 1990, pp. 363-368). As noted above, laboratory rats deficient in frontal lobe dopamine have a decreased ability to learn maze tasks (Comings, 1990, pp. 390-391).
Strong evidence also argues for a role for norepinephrine in attention deficits. First is that stimulant medications such as methylphenidate also have a strong stimulating effect on norepinephrine-responsive neurons. Second, norepinephrine is the major neurotransmitter of the locus cerulus, a part of the reticular activating system which regulates arousal and signal to noise ratio (Comings, 1990, pp. 409-412; Noback, Strominger, and Demarest, 1991, pp. 241-244; Ziegler and Lake, 1984, pp 110-111). Third, norepinephrine plays a role in regulating both dopamine and serotonin secreting neurons (Comings, 1990, pp. 411-413). Fourth, some symptoms of ADHD respond well to desipramine or to clonidine, two drugs which act primarily on norepinephrine responsive neurons.
Similar to some actions of norepinephrine, serotonin serves to regulate the activity of dopamine, especially by the influence of the raphe nuclei in the midbrain on the nucleus accumbens and striatum (Comings, 1990, pp. 446-447). However, different types of serotonin receptors (at least seven are known) have different functions which often compete with each other. To date, experience in treating ADHD with drugs active on serotonin receptors is limited, and no such drug has an FDA approved indication for the treatment of ADHD. However, anecdotal reports indicate that some ADHD patients respond favorably to fluoxetine (Prozac) or sertraline (Zoloft), though these drugs seem to affect more than one type of serotonin sensitive neuron.
In summary, various drugs which have activity on neurons sensitive to dopamine, and/or norepinephrine, and/or serotonin often have beneficial effects on ADHD. However, to date no purely neurochemical model fully explains ADHD, though future discoveries may uncover a purely biochemical cause for one or more types of ADHD. Present knowledge is more supportive of anatomic rather than biochemical causes of ADHD, though both are possible.
Hypothetical subtypes of ADHD
Present knowledge of the brain's attentional matrix, combined with other experimental data and clinical observations, suggest that there may be several subtypes of attention deficits. The following are brief clinical and electrophysiologic descriptions of subtypes which are theorized to correspond with the above described anatomic components of the attentional matrix.
A child with a weak reticular activating system might be impersistent in tasks, prone to mental errors, and in need of excessive stimulation to stay optimally aroused. Such a child might also have one or more specific learning disabilities caused by inadequate arousal of relatively weak cortical areas which have the potential to function in the normal range when fully aroused. The quantitative EEG of a "reticular-type" ADHD would exhibit high theta-beta ratios which are symmetrically distributed.
When frontal lobe dysfunction is the main cause of ADHD, one would expect a patient who is impulsive, hyperactive, and without organizational ability. He may also be rather oppositional, defiant, or emotionally labile; conversely, he may lack emotional expression. He may have problems orienting to novel stimuli, which may impact learning motivation. And, he may also have one or more specific learning disabilities, including dyslexia. The quantitative EEG of such a patient may show a tail of excessive theta activity in the left or, more probably, the right prefrontal area.
"Parietal type" ADHD might manifest as daydreaming and excessive distractibility. Such patients may also have specific learning disabilities, including dyslexia, disorders of visual processing (Demasio, 1985), mathematical processing (Benson and Geschwind, 1985, p. 224), and the comprehension of prosody (Ross, 1985, p. 248). On quantitative EEG, such patients would show a tail of excessive theta activity in the right parietal area.
A patient with a modality specific deficit would tend to be distractible by only one type of sensory input, such as visual images involving certain shapes or colors; he would also be learning disabled in the same sensory modality. Theoretically, such a patient may have a focal area of theta excess overlying the cortical area of the involved sensory modality.
A patient whose attention deficit is due to limbic dysfunction may seem to lack motivation, and may fail to orient or react to novel stimuli, or may respond inappropriately to inconsequential stimuli. He may also be hyperactive and may have problems with temporal sequencing. A patient with striatal dysfunction may be hyperactive or hypoactive, and may tend to overlook important stimuli. Patients with thalamic dysfunction can mimic any of the above clinical pictures, depending on which area of the thalamus malfunctions; in particular, they may have a disability of selective attention. And finally, the possibility exists that ADHD or one of its purported subtypes may represent some combination of malfunctions in two or more areas of the brain's attentional matrix.
Implications for Future Developments in Neurofeedback.
As possible subtypes of ADHD become elucidated and defined in detail, neurofeedback treatment of ADHD may adjust accordingly. Knowledge of subtypes of ADHD and their underlying neurophysiology may lead to modifications in lead placement and protocol. Better understanding of the interrelationships between certain specific learning disabilities and subtypes of ADHD may lead to enhanced treatment capabilities for specific learning disabilities. And finally, certain subtypes may be relatively more or less amenable to neurofeedback therapy; that is, the ability to recognize subtypes may enhance the effectiveness of neurofeedback by helping the clinician choose the most appropriate candidates.
Andreassi JL (1989). Psychophysiology: Human Behavior and Physiological Response, Second Edition. Lawrence Erlbaum Associates, Hillsdale, NJ.
Benson DF and Geschwind N (1985). Aphasia and related disorders: A clinical approach. In Mesulam, M-M, ed., Principles of Behavioral Neurology, F.A. Davis Company, Philadelphia, pp. 193-238.
Carpenter MB and Sutin J (1983). Human Neuroanatomy. Williams and Wilkins, Baltimore.
Comings DE (1990). Tourette Syndrome and Human Behavior. Hope Press, Duarte, CA.
Demasio AR. Disorders of complex visual processing: Agnosias, achromatopsia, Balint's syndrome, and related difficulties of orientation and construction. In Mesulam M-M, ed., Principles of Behavioral Neurology, F.A. Davis Company, Philadelphia, pp. 259-288.
Duffy FH, Denkla MB, Bartels PH, and Sandini G (1980). Dyslexia: Regional differences in brain electrical activity by topographic mapping. Annals of Neurology, 7:412-420.
Halperin JM, McKay KE, Matier K, and Sharma V, in press. Attention, response inhibition, and activity level in children: developmental neuropsychological perspectives. To appear in: Advances in Child Neuropsychology, vol 2, Michael G. Tramontana and Stephen R. Hooper, eds.
Knight RT. Electrophysiology in behavioral neurology (1985). In Mesulam, M-M, ed. Principles of Behavioral Neurology, F.A. Davis Company, Philadelphia, pp. 327-346.
Lou HC, Henriksen L, Bruhn P, Borner H, and Nielsen JB (1989). Striatal dysfunction in attention deficit and hyperkinetic disorder. Archives of Neurology, 46:48-52.
Lubar JF (1991). Discourse on the development of EEG diagnostics and biofeedback for attention deficit/hyperactivity disorders. Biofeedback and Self-Regulation, 16, 3:201-225.
Lynch JJ (1985). The Language of the Heart. Basic Books, Inc., New York.
Mann CA, Lubar JR, Zimmerman AW, Miller CA, and Muenchen RA (1992). Quantitative analysis of EEG in boys with attention-deficit/ hyperactivity disorder (ADHD): Controlled study with clinical implications. Pediatric Neurology, 8, 1:30-36.
Mesulam, M-M (1985-a). Patterns in behavioral neuroanatomy: Association areas, the limbic system, and hemispheric specialization. In Mesulam, M-M, ed. Principles of Behavioral Neurology. F.A. Davis Company, Philadelphia, pp. 1-70.
Mesulam, M-M (1985-b). Attention, confusional states, and neglect. In Mesulam, M-M, ed. Principles of Behavioral Neurology. F.A. Davis Company, Philadelphia, pp. 125-168.
Noback CR, Strominger, NL, and Demarest, RJ (1991). The Human Nervous System. Lea and Febiger, Philadelphia.
Ross ED. Modulation of affect and nonverbal communication by the right hemisphere. In Mesulam M-M, ed., Principles of Behavioral Neurology, F.A. Davis Company, Philadelphia, pp. 239-257.
Satterfield JH, Cantwell DP, Lesser LI, and Podosin RL (1972). Physiologic studies in the hyperkinetic child: I. American Journal of Psychiatry, 12, 11:1418-24.
Stuss DT and Benson DF (1986). The Frontal Lobes. Raven Press, New York.
Zentall S (1975). Optimal stimulation as theoretical basis of hyperactivity. American Journal of Orthopsychiatry 45, 4: 549-563.
Ziegler MG and Lake CR (1984). Norepinephrine. Williams and Wilkins, Baltimore, 1984.