The role of taxonomies in the study of human memory

The idea that memory is not unitary but is instead composed of multiple systems has a long history and has been debated with particular vigor in the last 20 years. Nevertheless, whether or not there are multiple memory systems remains unsettled. In this article, we suggest that psychologists wishing to classify memory can learn from biological systematics, the discipline that creates taxonomies of species. In so doing, we suggest that psychologists have made two assumptions in classifying memory: that features of memory are perfectly correlated, and that there is a straightforward mapping between taxonomy and theory. We argue that these assumptions are likely to be incorrect, but we also argue that there is a place for taxonomy in the study of memory. Taxonomies of memory are organizational schemes for data--they are descriptive, not explanatory--and so can inspire theory, although they cannot serve as theories themselves.

Motor learning of compatible and incompatible visuomotor maps

Brain imaging studies demonstrate increasing activity in limb motor areas during early motor skill learning, consistent with functional reorganization occurring at the motor output level. Nevertheless, behavioral studies reveal that visually guided skills can also be learned with respect to target location or possibly eye movements. The current experiments examined motor learning under compatible and incompatible perceptual/motor conditions to identify brain areas involved in different perceptual-motor transformations. Subjects tracked a continuously moving target with a joystick-controlled cursor. The target moved in a repeating sequence embedded within random movements to block sequence awareness. Psychophysical studies of behavioral transfer from incompatible (joystick and cursor moving in opposite directions) to compatible tracking established that incompatible learning was occurring with respect to target location. Positron emission tomography (PET) functional imaging of compatible learning identified increasing activity throughout the precentral gyrus, maximal in the arm area. Incompatible learning also led to increasing activity in the precentral gyrus, maximal in the putative frontal eye fields. When the incompatible task was switched to a compatible response and the previously learned sequence was reintroduced, there was an increase in arm motor cortex. The results show that learning-related increases of brain activity are dynamic, with recruitment of multiple motor output areas, contingent on task demands. Visually guided motor sequences can be linked to either oculomotor or arm motor areas. Rather than identifying changes of motor output maps, the data from imaging experiments may better reflect modulation of inputs to multiple motor areas.

Implicit motor sequence learning is represented in response locations

Previous work (Willingham, 1999) has indicated that implicit motor sequence learning is not primarily perceptual; that is, what is learned is not a sequence of stimuli. Still other work has indicated that implicit motor sequence learning is not specific to particular muscle groups or effectors. In the present work, we tested whether implicit motor sequence learning would be represented as a sequence of response locations. In Experiment 1, learning showed very poor transfer when the response locations were changed, even though the stimulus positions were unchanged. In Experiment 2, participants switched their hand positions at transfer, so that one group of participants pushed the same sequence of keys but used a different sequence of finger movements to do so, whereas another group pushed a different sequence of keys but used the same sequence of finger movements used at training. Knowledge of the sequence was shown at transfer only if the sequence of response locations was maintained, not the sequence of finger movements.

Effect of sex and joystick experience on pursuit tracking in adults

Using a joystick, adults (n = 39 males, 40 females in Experiment 1; n = 35 males, 40 females in Experiment 2; and n = 18 males, 18 females in Experiment 3) performed a computerized pursuit tracking task. Contrary to previously reported findings, the males were not more accurate than the females when performance was adjusted for prior perceptual-motor experience. Although no sex differences were found in a speeded tracking task, in an inverted tracking task the males exhibited a significant performance advantage; that advantage remained after several blocks of practice. Because participants' performance was adjusted statistically for prior perceptual-motor experience, the male advantage in inverted tracking was not related to experience. Rather, more proficient inverted tracking performance was associated with higher 3-dimensional mental rotations scores. In sum, sex differences in normal pursuit tracking may be better explained by differences in perceptual-motor experience. Inverted tracking, however, may depend on proficiency with spatial transformations.

Implicit motor sequence learning is not purely perceptual

Many reports have indicated that implicit learning of sequences in a choice response time task is primarily perceptual; subjects learn the sequence of stimuli rather than the sequence of motor responses. Three experiments tested whether implicit motor sequence learning could be purely perceptual: no support was found for that hypothesis. Subjects who merely watched stimuli did not learn the sequence implicitly (Experiment 1), and sequence learning transferred robustly to a different set of stimulus cues (Experiment 2). In the final experiment, the stimulus-response mapping was changed at transfer so that one group of subjects pushed the same sequence of keys but saw new stimuli, whereas another group pushed a different sequence of keys but saw the same stimuli. Transfer to the new mapping was shown only if the motor sequence was kept constant, not the perceptual sequence. It is proposed that subjects learn a sequence of response locations in this and similar tasks.

What differentiates declarative and procedural memories: reply to Cohen, Poldrack, and Eichenbaum (1997)

CPE claim that procedural and declarative representations differ on two important dimensions: flexibility and compositionality. I have proposed that the apparent flexibility of a memory depends entirely on the transfer conditions. Any retest is, in some sense, a test of flexibility, because something has changed since the original encoding episodic. I have argued that if one changes something that does not provide support to memory performance, the memory will appear flexible, and resistant to changes in the environment. If one changes the very thing that the representation codes, the memory will appear inflexible and easily disrupted by changes in the environment. This principle is equally true for procedural and declarative memory. CPE contend that procedural representations lack compositionality. An ideal test of this claim would examine the representation of a task that is widely agreed to be procedural (e.g. that has been demonstrated to be learned normally by amnesic patients, and in the absence of awareness by neurologically intact subjects). Such experiments appear not to have been conducted, and the fact is that many tasks that are widely agreed to be procedural probably are not compositional. They appear to be, as CPE contend, biases in a processing system; it is hard to imagine how repetition priming could be compositional. Nevertheless, this is not true of all procedural memories. There is a good deal of evidence that motor behaviour is organised hierarchically and has compositionality. There is every reason to think that most if not all motor behaviour is procedural; motor behaviour might be driven by goals that are declarative, but the low-level operations that actually manipulate effectors are closed to consciousness, do not depend on the medial temporal lobe or diencephalon, and would therefore be classified as procedural. CPE framed their theory of differences between procedural and declarative memory systems as an account of the deficit in amnesic patients. They therefore predict that the learning of amnesic patients should not show flexibility or compositionality. There is already at least one study showing learning in amnesic patients that is as flexible as that of control participants (Knowlton & Squire, 1996). There are not, to my knowledge, data on whether the motor skill learning of amnesic patients shows compositionality, but one might expect that it would, given that it does in neurologically intact participants, and given that motor skill learning appears unimpaired in amnesic patients. Thus, the conception of declarative and procedural memory provided by CPE may not provide a complete account of amnesic performance. The anatomic distinction between procedural and declarative memory systems appears quite strong, and there is therefore reason to believe that there are accompanying computational differences. There does not, however, appear to be sufficient evidence to support those differences proposed by CPE.

A neuropsychological theory of motor skill learning

This article describes a neuropsychological theory of motor skill learning that is based on the idea that learning grows directly out of motor control processes. Three motor control processes may be tuned to specific tasks, thereby improving performance: selecting spatial targets for movement, sequencing these targets, and transforming them into muscle commands. These processes operate outside of awareness. A 4th, conscious process can improve performance in either of 2 ways: by selecting more effective goals of what should be changed in the environment or by selecting and sequencing spatial targets. The theory accounts for patterns of impairment of motor skill learning in patient populations and for learning-related changes in activity in functional imaging studies. It also makes a number of predictions about the purely cognitive, including accounts of mental practice, the representation of motor skill, and the interaction of conscious and unconscious processes in motor skill learning.

A neuropsychological theory of motor skill learning

This article describes a neuropsychological theory of motor skill learning that is based on the idea that learning grows directly out of motor control processes. Three motor control processes may be tuned to specific tasks, thereby improving performance: selecting spatial targets for movement, sequencing these targets, and transforming them into muscle commands. These processes operate outside of awareness. A 4th, conscious process can improve performance in either of 2 ways: by selecting more effective goals of what should be changed in the environment or by selecting and sequencing spatial targets. The theory accounts for patterns of impairment of motor skill learning in patient populations and for learning-related changes in activity in functional imaging studies. It also makes a number of predictions about the purely cognitive, including accounts of mental practice, the representation of motor skill, and the interaction of conscious and unconscious processes in motor skill learning.

Deficit in learning of a motor skill requiring strategy, but not of perceptuomotor recalibration, with aging

We investigated the effect of aging on different aspects of motor skill learning using two computer-presented perceptuomotor tasks. The relationship between visual and proprioceptive feedback was transformed in the first task, which was open to the formation and use of strategies. This task was designed to lead to perceptuomotor adaptation that was then measured by performance on a very similar second task that was not open to the use of strategy task. Older participants showed impaired learning of the strategic task but not of the nonstrategic task. This is in line with the suggestion that the effect of aging on learning and memory may be to reduce working memory resources.

Response-to-stimulus interval does not affect implicit motor sequence learning, but does affect performance

Nissen and Bullemer (1987) reported that implicit motor sequence learning was disrupted by the addition of a secondary task. They suggested that this effect was due to attentional load that the secondary task adds. Recently it has been suggested that the attentional load is not critical, but rather that the secondary task affects timing, either by lengthening or by making inconsistent the response-to-stimulus interval (RSI)--that is, the delay between when a subject makes a response and when the next stimulus appears. In six experiments we manipulated the RSI and found no support for these two hypotheses. An inconsistent RSI did not adversely affect implicit motor sequence learning. A long RSI did not affect learning, although under some conditions subjects did not express learning if the RSI was long. These results are interpreted as reflecting the effects of attention.

Intact mirror-tracing and impaired rotary-pursuit skill learning in patients with Huntington's disease: evidence for dissociable memory systems in skill learning

Skill learning in early-stage Huntington's disease (HD) patients was compared with that of normal controls on 2 perceptual-motor tasks, rotary pursuit and mirror tracing. HD patients demonstrated a dissociation between impaired rotary-pursuit and intact mirror-tracing skill learning. These results suggest that different forms of perceptual-motor skill learning are mediated by separable neural circuits. A striatal memory system may be essential for sequence or open-loop skill learning but not for skills that involve the closed-loop learning of novel visual-response mappings. It is hypothesized that working memory deficits in HD resulting from frontostriatal damage may account broadly for intact and impaired long-term learning and memory in HD patients.

Intact mirror-tracing and impaired rotary-pursuit skill learning in patients with Huntington's disease: evidence for dissociable memory systems in skill learning

Skill learning in early-stage Huntington's disease (HD) patients was compared with that of normal controls on 2 perceptual-motor tasks, rotary pursuit and mirror tracing. HD patients demonstrated a dissociation between impaired rotary-pursuit and intact mirror-tracing skill learning. These results suggest that different forms of perceptual-motor skill learning are mediated by separable neural circuits. A striatal memory system may be essential for sequence or open-loop skill learning but not for skills that involve the closed-loop learning of novel visual-response mappings. It is hypothesized that working memory deficits in HD resulting from frontostriatal damage may account broadly for intact and impaired long-term learning and memory in HD patients.

Patients with Alzheimer's disease who cannot perform some motor skills show normal learning of other motor skills

Previous researchers have claimed that patients with Alzheimer's disease (AD) learn new motor skills normally, although many AD patients cannot perform the tasks and must be eliminated from the analysis. Excluding them assumes that they have a deficit of motor performance (competence to perform the task), but not of motor learning (ability to improve performance). The present study administered 4 motor tasks to 20 AD patients and 20 controls. The results showed that the ability to complete 1 task (performance) did not predict the rate of improvement (learning) on another task, which indicates that AD patients do indeed have a performance deficit and not a general deficit of motor skill learning. Dementia ratings predicted the ability to perform tasks but not the ability to learn them. It is concluded that it is defensible to claim that AD patients learn a motor skill normally, even if some of the patients are unable to perform the task.

On the development of procedural knowledge

Amnesic patients demonstrate by their performance on a serial reaction time task that they learned a repeating spatial sequence despite their lack of awareness of the repetition (Nissen & Bullemer, 1987). In the experiments reported here, we investigated this form of procedural learning in normal subjects. A subgroup of subjects showed substantial procedural learning of the sequence in the absence of explicit declarative knowledge of it. Their ability to generate the sequence was effectively at chance and showed no savings in learning. Additional amounts of training increased both procedural and declarative knowledge of the sequence. Development of knowledge in one system seems not to depend on knowledge in the other. Procedural learning in this situation is neither solely perceptual nor solely motor. The learning shows minimal transfer to a situation employing the same motor sequence.

On the development of procedural knowledge

Amnesic patients demonstrate by their performance on a serial reaction time task that they learned a repeating spatial sequence despite their lack of awareness of the repetition (Nissen & Bullemer, 1987). In the experiments reported here, we investigated this form of procedural learning in normal subjects. A subgroup of subjects showed substantial procedural learning of the sequence in the absence of explicit declarative knowledge of it. Their ability to generate the sequence was effectively at chance and showed no savings in learning. Additional amounts of training increased both procedural and declarative knowledge of the sequence. Development of knowledge in one system seems not to depend on knowledge in the other. Procedural learning in this situation is neither solely perceptual nor solely motor. The learning shows minimal transfer to a situation employing the same motor sequence.

Memory and awareness in a patient with multiple personality disorder

We studied an individual with multiple personality disorder in whom each of several personalities claimed to have no direct awareness of the others and to be unable to consciously remember the experiences of other personalities. A broad selection of implicit and explicit memory tests was used to determine the extent to which one personality had access to knowledge acquired by another and the circumstances in which that knowledge would be expressed. The implicit assessment of memory was a necessary but not sufficient condition for demonstrating interpersonality access. The degree of compartmentalization of knowledge in this patient depended largely on whether the interpretation of presented information was likely to differ across personalities.