|Home | About | Journals | Submit | Contact Us | Français|
The present review develops a framework from which to understand the role of the cholinergic system in healthy cognition and in cognitive dysfunction. Traditionally, the cholinergic system has been thought to have direct influence on cognitive processes such as working memory and attention. Although the influence of cholinergic function on stimulus processing has been long appreciated, the notion that cholinergic effects on stimulus processing is the mechanism by which acetylcholine influences cognitive processes has only more recently been considered.
Literature supporting the hypothesis that cholinergic modulation influences cognitive functions through stimulus processing mechanisms has been growing for over a decade. Recent conceptualizations of the developing literature have argued for a new interpretation to an old and developing literature.
The argument that cholinergic function modulates cognitive processes by direct effects on basic stimulus processing extends to cognitive dysfunction in neuropathological conditions including dementia and mood disorders. Memory and attention deficits observed in these and other conditions can be understood by evaluating the impact of cholinergic dysfunction on stimulus processing, rather than on the cognitive function in general.
Acetylcholine was the first neurotransmitter identified in the human body and, thus, was the first to undergo major study. The culmination of decades of work point primarily to the role of acetylcholine in stimulus processing and memory/attention, and current interests include efforts to understand how the dysfunction of this system contributes to conditions and syndromes that implicate these basic cognitive functions. The goal of this review is to offer a cognitive framework based on our understanding of basic cholinergic function from which to better understand cholinergic dysfunction and cognition in human illness.
Cells that constitute the cholinergic neurotransmitter system originate primarily in the basal forebrain. These cells provide extensive projections throughout the cortex and, thus, by design have widespread effects on information processing. Cholinergic neuromodulation is known to influence multiple cognitive processes, including memory and attention [1,2]. In rodents and in nonhuman primates, lesions of the cholinergic nucleus basalis of Meynert (nBM)  result in impaired performance on learning and memory tasks. Blockade of the cholinergic muscarinic receptors by the antagonist scopolamine results in impaired performance on learning and memory tasks [4,5,6••,7•,8] and on tasks of attention [1,9], whereas enhancing cholinergic function improves memory and attention . In healthy humans, scopolamine produces a transitory impairment of a wide range of memory and attention functions [11–14]. In contrast, several drugs that enhance cholinergic neuromodulation improve performance on short-term memory tasks both in animals and in humans [13,15–17] and can reverse the memory deficits created by nBM lesions . Historically, the literature has identified acetylcholine as related to cognition and cognitive processing. A closer look at the literature, however, argues that cholinergic effects on cognitive functions, such as working memory and attention, occur specifically as a result of direct effects on stimulus processing mechanisms.
The literature is rich with evidence of the involvement of the cholinergic system in memory and attention mechanisms [1,18–24]. Researchers have hypothesized that attentional processes are mediated through cholinergic mechanisms that facilitate the processing of sensory information [13,24,25] and evidence exists to support this idea [13,26,27].
In general, the cholinergic neurons of the basal forebrain that project throughout neocortex are thought to enhance signal-to-noise ratios (S/N) for neural sensory processing [24,28,29]. Early studies by Sillito and Kemp  and Sato et al.  demonstrated that the direct application of acetylcholine to cat visual cortex increased the selectivity of the cells, response to stimulus orientation, by enhancing response selectivity and increasing response magnitude, consistent with the hypothesis that acetylcholine modulates S/N. Similarly, Buzsaki  showed that cholinergic input to hippocampus is inhibitory, suggesting that acetylcholine may enhance S/N in hippocampus by reducing the response to noise. More recently, comprehensive animal work by Sarter et al.  and Hasselmo and Sarter  provides convincing evidence that the role of the cholinergic system in sensory information processing that leads to stimulus or cue detection, via S/N modulation, is central to cognitive functioning.
Human studies using functional brain imaging techniques designed to evaluate the effects of cholinergic modulation on cognitive functions such as working memory and attention have identified changes in neural activity that are consistent with modulation of stimulus processing [13,32]. Direct modulation of S/N may constitute the neural mechanism through which the cholinergic system may establish the relative strengths of the neural representations of stimuli. For example, a functional imaging study  demonstrated that enhanced cholinergic activity selectively increased neural responses to task-relevant stimuli (i.e. signal) with reduced or no change in neural responses to task-irrelevant stimuli (i.e. noise), consistent with a selective enhancement for target stimuli via S/N processing. These findings showing stimulus-specific effects of cholinergic modulation are consistent with the hypothesis that cholinergic activity influences cognitive processes by influences on stimulus processing mechanisms.
Working memory denotes a cognitive process that temporarily maintains an active representation of information for further processing or recall [33,34]. The cholinergic system strongly modulates working memory, whereby enhancing cholinergic activity improves working memory [35–38] and blocking normal cholinergic function impairs working memory performance [5,8,39]. Functional brain imaging studies have facilitated the understanding of neural mechanisms that underlie cholinergic effects on cognitive function [6••,9–15].
Functional brain imaging studies have demonstrated that increases in cholinergic activity preferentially enhance neural responses [26,40,41] and blocking cholinergic function reduces neural responses selectively during stimulus task encoding processes [42,43]. In a series of studies with PET and functional magnetic resonance imaging (fMRI), enhancing cholinergic function with an anticholinesterase modulated neural response to a working memory task across brain regions [17,26,36–38] but the only regions consistently found to show an increase in neural activity included visual processing areas, consistent with the idea that cholinergic function directly influences stimulus processing given the visual nature of the stimuli. Moreover, an fMRI study that evaluated each working memory component (i.e. encoding, maintenance, recognition) demonstrated increases in neural response selectively to task-relevant stimuli in ventral visual cortical regions, particularly during stimulus encoding . Others also have reported selective effects during the encoding phase of working memory [32,40,41], a finding that is consistent with the hypothesis that improvement in working memory following cholinergic enhancement is mediated by influences on stimulus processing. Early on, these findings were considered paradoxical  in that enhanced working memory function was expected to be associated with enhancement of classic working memory regions in prefrontal cortex. The absence of enhanced prefrontal cortical function as demonstrated with functional imaging, together with the isolated enhancement of function in stimulus processing regions, pointed to a cholinergically mediated modulation of basic stimulus processing mechanisms in the context of a working memory task.
Selective attention constitutes the ability to discriminate significant or relevant stimuli from irrelevant stimuli (i.e. noise) and to process information preferentially [44–47]. The presentation of multiple stimuli simultaneously produces a competition for neural representation [44,48]. Single-unit recording studies [44,49,50] and functional brain imaging studies [48,51,52] have demonstrated that the processing of a visual stimulus is influenced by the presence of other, unattended visual stimuli. Two mechanisms that each contribute to the biasing of attention, including ‘bottom-up’ and ‘top-down’ processes [44,53], are thought to resolve the competition among stimuli. Bottom-up stimulus-based mechanisms refer to neural processing that is biased toward stimuli with inherently salient or meaningful features (i.e. stimuli that retain sensory salience, or that hold biological relevance) [54–57]. Top-down mechanisms refer to knowledge-based processes in which attention is oriented intentionally, resulting in the enhancement of neural representations of relevant, goal-directed stimuli [58,59]. The interaction of bottom-up and top-down mechanisms [58,60,61] produces a biased neural representation of the stimuli. Both bottom-up and top-down processing mechanisms are mediated through the cholinergic system and, thus, selective attention reflects the combined influence of cholinergic processing effects via these mechanisms (reviewed by ).
In a behavioral study of selective attention, in which two stimuli were presented simultaneously (face and house) and, thus, competed for representation, the cholinergic system was both enhanced using physostigmine and blocked using scopolamine . Participants were instructed to attend to one stimulus component (face or house) and perform a matching task while ignoring the other stimlus. The expectation was that a processing bias toward one stimulus type over another would be reflected in performance measures and would be altered by cholinergic manipulation. For example, a face holds more innate salience and will be easier to attend to (while ignoring the house) and harder to ignore (when attending to the house). The results showed that the effects on performance measures were selective to the attention/target stimulus condition both during cholinergic enhancement and impairment, indicating that the effects of cholinergic modulation are stimulus-dependent. Specifically, enhancing cholinergic activity resulted in a selective reduction in reaction time when attending to houses (thus reducing the baseline bias toward faces) and impairing cholinergic activity increased reaction time selectively when attending to faces (also reducing the bias toward faces). The findings are consistent with the hypothesis that the behavioral outcome reflects the resolution of stimulus interactions via top-down and bottom-up mechanisms, and that the effects are driven by stimulus properties that contribute to the resolution of this competition.
The extent to which cognitive impairment is explained by cholinergic dysfunction in neurological and psychiatric conditions remains empirical. As dysregulation of any single neurotransmitter system results in cascading effects throughout other transmitter systems, attributing behavioral deficits to single transmitter systems is challenging. Nonetheless, several neurological and psychiatric conditions that have hallmark neurocognitive features and implicate the cholinergic system will be discussed.
Aging is associated with anatomical, chemical, and functional changes in the brain [62,63], the most prominent of which are alterations in the cholinergic system. These changes include decreases in basal forebrain cholinergic neurons, cholinergic receptors, and afferent projections to cortex [64–66]. Although their significance has been questioned [63,67], these age-associated changes have led to the cholinergic hypothesis of aging, which suggests cholinergic alterations contribute to the deficits in working memory, attention, and other cognitive functions observed during aging [25,62,68,69•]. The pharmacological potentiation of cholinergic neurotransmission improves performance on cognitive tasks in the elderly [37,38,70], and chronic treatment with drugs that enhance the cholinergic function is used to ameliorate cognitive dysfunction in the elderly [71,72].
The study of working memory in aging [73,74] indicates that the elderly recruit brain regions (including prefrontal cortical regions) that are not recruited in younger individuals performing the same working memory task, a finding that is thought to reflect compensatory mechanisms that accompany aging . In a simple delayed-match-to-sample working memory task used in groups of young and healthy elderly individuals, cholinergic enhancement reduced activity in the prefrontal regions that had been selectively recruited during working memory in each age group. Notably, visual cortical areas were the only brain regions to show increases in activity during cholinergic enhancement and working memory in both groups, and this increase was larger in older participants . In a similar working memory study with a group of young and a group of older healthy participants, task delay was modulated to increase task difficulty. Again we observed reduced prefrontal activity across levels of task difficulty during cholinergic enhancement, and visual processing areas showed increases in activity . Moreover, elderly participants recruited more extended visual processing areas than did younger participants, but both groups showed increased neural activity exclusively in visual processing areas. Together, these findings suggest that even in the presence of changes in the cholinergic system and compensatory functional changes in aging, cholinergically mediated improvements in working memory performance appear to be associated with augmentation of stimulus processing mechanisms.
A recent paper evaluating aging and memory postulates a specific neural mechanism responsible for memory impairment. Specifically, young and older adults showed reduced activity in brain regions important for encoding when encoding was unsuccessful, but older individuals also showed increased activity in regions that likely mediate distraction. This argument is consistent with changes in S/N processing, with an increase in noise overwhelming the signal, and thus fits well with the hypothesis that cognitive dysfunction in aging is associated with changes in stimulus-based processing mechanisms.
Alzheimer’s disease implicates many neurotransmitter systems and includes widespread cortical damage, but the cholinergic system is the most prominently implicated system. Modulation of cholinergic activity improves cognitive impairments [76•] and affects task-specific neural activity in Alzheimer’s disease . Although cognitive function is impaired across multiple domains, working memory shows deficits early in the disease process . In another working memory study, a task that included the variation in task difficulty was administered to patients with early Alzheimer’s disease to evaluate the impact of enhancing cholinergic function on task performance. In a group of five patients, significant improvement in performance accuracy was observed in the longer working memory delay conditions under cholinergic enhancement as compared to placebo (Fig. 1; unpublished finding), with accuracy increasing from approximately 62% during the longest working memory delay under placebo, to approximately 95% under physostigmine. On the basis of the results reported from both healthy young and healthy older participants discussed above, we can argue that the improvement in task performance in patients suffering from Alzheimer’s disease likely results from the enhancement of stimulus processing in early visual areas, and by extension the working memory impairment may be due at least partially to deficits in stimulus processing mechanisms rather than in the cognitive aspects of working memory per se.
Multiple neurotransmitter systems also are implicated in depressive disorders including the dopaminergic, serotonergic, noradrenergic, and cholinergic systems. The cholinergic system has been found to be hypersensitive in depression, whereby depressed patients show exaggerated neuroendocrine and pupillary responses to cholinergic agents [79–81] and experience forms of sleep disturbance (decreases latency to REM and increased REM density) that are consistent with increased cholinergic muscarinic sensitivity [82,83]. Janowsky et al. [84–86] reported that in manic bipolar patients, increasing cholinergic activity induced depressive symptoms, and in major depressive disorder (MDD), increasing cholinergic activity worsened symptoms of depression [87–89]. The role of the cholinergic system in mood disorders has been highlighted more recently through the demonstration that blocking cholinergic muscarinic activity with scopolamine produces rapid antidepressant effects [90•,91].
Behavioral and cognitive features of depression are associated primarily with the processing of affective information. A consistently reported finding is a mood congruent processing bias in depressed individuals, which is defined as a tendency to show a bias for processing negative information as compared to positive or neutral information [92,93,94••,95]. Results of memory studies show that MDD patients recall more negatively toned material than positively toned information [96–99]. In the context of attention paradigms [100–102], the influence of mood congruent processing is demonstrated by the finding that depression-related words produce more interference on emotional stroop tasks than do happy or neutral words. Similarly, Murphy et al.  showed in an affective attention-shifting task that depressed individuals are slower in responding to the presentation of happy word as compared to sad word targets, and that their ability to shift attention from happy to sad or sad to happy targets is impaired.
The mood congruent processing bias observed in MDD readily can be characterized within the framework of the cholinergic system and stimulus processing mechanisms. The biased processing of negative or sad information is consistent with an overactive cholinergic system in depression resulting in the over-representation of negative information. This framework would hypothesize that competition among competing stimuli in the environment engages the cholinergic system, and the overactive system alters the bias preferentially toward negative stimuli in MDD. A functional brain imaging study that used a selective attention task with images of emotional faces and houses observed processing biases between emotional faces in visual processing areas that were opposite to each other in healthy controls and patients with MDD , a finding that would be predicted by this hypothesis. The effect of cholinergic modulation on these baseline differences will be informative.
The cholinergic neurotransmitter system traditionally has been linked with cognitive functions including attention and memory. Evidence concerning the direct effects of cholinergic function on stimulus processing, together with findings from cognitive studies that characterize stimulus processing effects within the context of cognitive functions, leads to the hypothesis that the cholinergic system retains the role of supporting and modulating the processing of task-related stimuli in the context of cognitive functions. This concept carries forward to pathological conditions that both implicate cholinergic activity and retain hallmark cognitive features that can be explained by changes in cholinergic influence on stimulus processing.
The author would like to thank the many colleagues who have collaborated on various aspects of work that is reflected in this review, with particular thanks to Drs James Haxby, Pietro Pietrini, and Emiliano Ricciardi. The author would also like to thank Drs Emiliano Ricciardi and Allison Nugent for comments on the article. The work presented here associated with the author was supported by the intramural research programs of the NIA/National Institutes of Health and the National Institute of Mental Health/NIH. A use-patent application for the use of scopolamine as an antidepressant agent has been filed by the NIMH/NIH.
Papers of particular interest, published within the annual period of review, have been highlighted as:
• of special interest
• of outstanding interest
Additional references related to this topic can also be found in the Current World Literature section in this issue (p. 411).