Data collected during the past years indicate that adult auditory cortex (including the primary cortex) is capable of undergoing substantial plasticity. The most prominent change, found under different behavioral conditions and using a variety of recording techniques, is a specific increase in the extent of cortical representation of the experimentally enhanced range of tone frequencies: a fast establishment of a conditioned fear response to pure tone is accompanied by a fast [1•], long-lasting [2•] shift of neuronal characteristic frequencies towards the frequency of the conditioned stimulus. Consequently, the representation of this frequency (measured with fluorodeoxyglucose) is increased [3]; improvement in (pure tone) frequency discrimination is correlated with a long-term increase in the representation of the whole range of stimulating frequencies [4••] — although training to discriminate between spectral envelopes mainly sharpens neuronal tuning curves (D Keeling, K Krueger, B Calhoun, CE Schreiner, abstract 17:22, Assoc Res Otolaryngol, St. Petersburg Beach, Florida, February 1994). Frequency-specific deafferentation also induces short- and long- term expansion of the cortical representation of spared neighboring frequencies [5][6][7•].

The direct behavioral consequences of the extended representation have not been extensively studied. The nearly ubiquitous expansion (although see [8•]) is puzzling and raises several questions as to its cause, on the one hand, and its consequences, on the other. For example, is it induced by a common local rule of plasticity operating at the intercellular level? Is it necessary and/or sufficient for behavioral improvement?

Local rules of plasticity

Increased cortical representations of experimentally enhanced input frequencies can be explained by learning mechanisms based on the Hebbian principle operating at the intercellular level. The original Hebb rule [9], combined with Stent's extension [10], suggests that a positive contingency between the increased activities of the pre- and postsynaptic neurons will strengthen their connection, whereas a negative contingency will weaken it.

According to this rule, input synapses conveying the enhanced range of inputs will strengthen, given that the postsynaptic cell is activated by these stimuli. Quiescent synapses, exhibiting in this case negative correlation with the output cell, will weaken. Consequently, tuning curves of postsynaptic cells will shift in favor of the enhanced stimulus range expanding its representation (for further discussion see, e.g. [11][12•][13•]).

However, none of the experiments demonstrating tuning shifts measured intercellular temporal contingency directly. Several studies (in visual and motor cortices) manipulated local contingency between a postsynaptic unit and an external or thalamic stimulation, applying invasive techniques [14][15][16][17][18][19][20]. These studies were highly informative in indicating that temporal contingency was necessary to induce modulations. However, as the effect of the intracranial stimulation was not well defined, and other synapses, e.g. cholinergic and noradrenergic, may have also been stimulated, these studies could not determine whether temporal contingency was sufficient. Furthermore, not having recorded a specific pre- and postsynaptic neuronal pair, a quantitative study of the required temporal contingency was not possible.

Simultaneous recording of two neurons in the behaving mammal is possible only when recorded extracellularly. A quantitative analysis of the nature of correlated activity yielding neuronal plasticity can be obtained applying cross-correlation analysis to the extracellularly recorded activities. This correlation reflects the statistical efficacy of the coupling, induced by direct, but also by indirect, synaptic connectivity. It is therefore termed functional coupling [21]. It can be shown that functional coupling describes activity covariance, mutual information and surprise functions. The competence of these functions in describing neural coding was supported by both theoretical accounts [22][23][24][25][26] and physiological findings [23][27][28][29][30].

A novel approach utilizes measures of lasting changes in the functional coupling to determine functional plasticity [27]. Plastic changes of functional couplings serve as an estimate of neuronal plasticity. Using this approach it was shown that functional plasticity is indeed guided by changes in the temporal contingency between the activities of the coupled cells: when conditioning increased the temporal covariance, the coupling was potentiated, whereas when it decreased the covariance, the coupling was depressed [27]. When the covariance was not affected by the conditioning (e.g. during pseudo-conditioning), the strength of the coupling was, in most cases, not changed (Fig. 1a).

Return to text reference [1] [2] [3] [4]

These findings partially support the covariance formulation of temporal contingency [22]. However, functional coupling was modified by the change in covariance and not by its absolute value (see also [31]). Thus, the reference level around which coupling was strengthened or depressed was the preconditioning, steady-state level and not zero covariance (expected only when there is no coupling). Such a reference level is beneficial as it overcomes the 'runaway' problem [32][33][34][35] embedded in the covariance formulation: excitatory synapses, inducing positive covariance, would be indefinitely facilitated towards some saturation value.

Learning rules that address the runaway problem incorporate a negative feedback component: changes of synaptic weights are inversely related to either the present synaptic weights or to the expected postsynaptic activities (reviewed in [34], and for classical conditioning in [35][36][37]). Negative feedback operating on the postsynaptic activity was in general formulated as follows:

w = g(x)c (y - < y >),.

where w is the change in synaptic weight, x and y denote the activities of the pre- and postsynaptic neurons, respectively, c is a constant, g is a monotonic function and < y > denotes the expected value of y.

The straightforward prediction of this formulation is that synaptic modifications will be correlated with changes in postsynaptic activities.

Negative feedback operating on the synaptic weight was generally formulated as follows [34]:

w = [g(x)c - w]h(y),.

where w is the synaptic weight and h is a monotonic function.

This formulation predicts, firstly, that there should be a negative correlation between pre-conditioning weights and the magnitude of changes induced by conditioning, and, secondly, that at steady-state, there should be a positive correlation between weights and presynaptic activity levels.

None of these three predictions was confirmed by analysis of functional plasticity [27][38]. The finding that weights were not correlated with activities of either the pre- or postsynaptic neuron, and weights changes were not correlated with changes of either activity, suggests that the negative feedback operates on a different factor. This factor may be the correlated activity itself, as in the following form:

w = g(y - < y >),.

where denotes a 'trace' of the pre-synaptic activity [35] on a scale of a few milliseconds, and < > denotes averaging over a longer time window, on the scale of a few minutes.

Nonetheless, a multiplicative dependency should also be considered, as it is suggested by experimental findings that the fractional change, rather than the differential change, was negatively correlated with the pre-conditioning strength [38]:

w/w = g(y/ < y >) -1.

The function g can be approximated by experimental data, e.g. see Fig. 1. The actual time windows used by cortical neurons for averaging activities or correlation of activities are yet to be determined.

Note that we tested rules derived for synaptic weights using results of functional couplings. While it is clear that these two measures should be strongly related, this testing could be refined either by deriving predictions of models in the functional domain or by translating these findings to the synaptic domain.

Intracellular normalization

Neuronal plasticity may also be the outcome of intrinsic cellular mechanisms [39]. Based on theoretical grounds, it was suggested that there should be intracellular mechanisms that are directed at maintaining stability of mean firing rates under changing input conditions, by modifying synaptic weights [40] or by changing excitability [39]. Physiological findings support the suggestion that neuronal excitability may change as a consequence of cellular or behavioral conditioning [41]. However, whether these mechanisms are aimed at maintaining output stability is questionable.

Mechanisms that stabilize the firing rate may explain results from behavioral studies of adaptation and input deprivation. For example, adaptation to frequency motion biases judgment towards the opposite direction [42•], whereas artificial scotoma biases spatial judgments towards the scotoma [43•]. Behavioral biases, induced by both conditions, may derive from misinterpretation of a 'readout' process resulting from modifications of activities of some of its inputs, which outlast stimulus manipulation. The modified inputs, in this case, are neurons that actively avoid extreme conditions of hyper- or hypoactivity and, thus, seek to maintain some output stability.

Locus of changes

Changes exhibited at the auditory cortex may be the consequence of changes within subcortical circuits, thalamocortical projections [44], or cortical, local [45•] or remote [46] connections. Naturally, these are not mutually exclusive options. In fact, recent evidence indicates that conditioning induces modifications at various areas along the lemniscal and extra-lemniscal path, including the ventral and dorsal cochlear nuclei [3][47•][48•]. Yet, the finding that subcortical stages modify does not indicate that modifications found at cortical level only reflect these peripheral changes. Indeed, in some cases, non-monotonous degrees of changes are found along the auditory pathway [3][48•]. Furthermore, at least under some conditions, cortical changes can be induced faster than subcortical changes [49].

It is not clear whether modifications at the various stages are subject to the same behavioral control (or local rule of plasticity). For example, different levels may be subject to similar local factors but to different types of behavioral control: for bottom-up induced plasticity, the primary focus of plastic changes may be subcortical, whereas for top-down induced modulations, the primary focus is expected to be cortical. Such a distinction was in fact implied in suggestions that deafferentation results mostly in subcortical changes (as in [50•][51], but perhaps not in [52]; see [53] for a recent review), whereas behavioral training leads to cortical and/or thalamo-cortical modifications ([4••], reviewed in [54•]). (The allocation of classical conditioning to any of these two categories probably depends on the specific paradigm employed.) In particular, it was suggested that cortical plasticity might be important in inhibition of learned reflexive responses [55••], difficult frequency discrimination [4••][55••], attentive pattern discrimination [4••] or storage of records of events for 'off-line' use [55••].

Additional support for the suggested anatomical segregation comes from the analysis of response latencies. Strengthening of connections along the input pathways is expected, on the average, to decrease the minimal latency in cortical responses to preferred stimuli, whereas weakening of these connections is expected to increase it. Such an increase was indeed found following training on frequency discrimination [4••], but not following cochlear lesions [50•]. This behaviorally induced cortical re-tuning can be explained by relative strengthening of cortical connections on account of thalamo-cortical connections (by means of intrinsic regulating, or normalizing, mechanisms), so as to preserve output stability. In that case, cortico-cortical and thalamo-cortical connections will co-change, but generally in opposite directions.

Computational considerations

As long as cortical algorithms and representations are not known, their optimization by learning cannot be deduced (see discussion in [56]). However, some general considerations are still valid. Unsupervised Hebbian local mechanisms can enhance mean responses of single cells and also increase the coherence between activities of different neurons participating in the same processing assembly (e.g. [57]). Both types of changes will lead to behavioral improvement by means of increased reliability and shortening of required processing time.

However, the most abundant change, presumably also related to unsupervised Hebbian mechanisms, namely expansion of representation, will not automatically lead to behavioral improvement. Obtaining behavioral benefits requires accompanying, probably supervised (i.e. directed by internal knowledge), update of the readout mechanism. Otherwise the induced shift of tuning curves will be misleading and not profitable, as is the case following adaptation or input deprivation. In fact, improvement may be achieved by task-specific optimization of the readout mechanism even without any changes in basic architecture [58•]. The implication that improvement requires task-specific behavioral control, even though architectural plasticity does not, is supported by behavioral findings [59•].

The suggestion that shifting tuning curves towards experimentally enhanced frequencies reflects unsupervised Hebbian mechanisms (probably influenced by general apriori behavioral relevance), whereas additional supervised mechanisms are involved in task- related processes, may account for apparent inconsistencies between studies of differential aversive conditioning. In one line of studies (reviewed in [60]), neurons increased their responses to the conditioned stimulus frequency, whereas in another study, neurons increased responses to frequencies neighboring the conditioned stimulus but not to the conditioned stimulus frequency [8•]. This difference may reflect two different processes activated with each stimulus presentation: an initial unsupervised, stimulus-driven process, and a subsequent, behaviorally guided, supervised process. Indeed the first type of modification was found studying initial 'on' responses and the latter when longer delays were analyzed (the first 250 ms). Interestingly, 'on' and 'sustained' response components were often found to be modified independently in the auditory cortex by either external stimuli [61] or cholinergic stimulation [62][63].

The suggestion that improvement of behavioral discrimination requires mechanisms operating with 100-200 ms latency from stimulus application, is supported by an EEG (electroencephalogram) auditory study [64••]. Practice-induced improvement of discrimination between a standard and a deviant complex auditory stimuli was correlated with an increase in the EEG component of 100–200 ms, originating probably from the auditory cortex [65].

Behavioral aspects

Behavioral training induces modality-specific modulations in receptive field properties [4••]. However, the extent of this specificity (e.g. the degree of channel- specificity) has not yet been studied. Furthermore, the relative significance of each of a variety of behavioral factors (e.g. apriori attention, stimulus-response reinforcement, and internal feed-back supervision) forming a behavioral context, in affecting neuronal plasticity, has not yet been studied. Modulations found following deafferentation while the animal is still anaesthetized, imply that under extreme conditions behavioral control may not be required (reviewed in [54•]), suggesting that its role, for at least some plastic changes, is modulatory and not permissive.

The importance of behavioral context for neuronal plasticity was directly demonstrated at the intercellular level, studying functional plasticity [27]. Potentiations obtained by cellular conditioning when a monkey was behaving were on average six times stronger than those obtained by a similar conditioning when the monkey was not behaving (change of 180% versus 30%; Fig. 1aFig. 1b). Intriguingly, this ratio fits Thorndike's estimate that ''a single occurrence followed by reward strengthens a connection about six times as much as it would be strengthened by merely occurring'' [66]. Although few neuronal pairs showed marked plasticity even under the non-behaving condition, all neuronal pairs recorded under both conditions exhibited stronger changes when the monkey was behaving.

The behavioral control of plasticity may be mediated by some (or all) of the ascending diffuse systems, and, in particular, the cholinergic and noradrenergic ones. These systems could be activated under different behavioral conditions and consequently act on cortical synapses. Recent studies of the cholinergic system have shown that increasing acetylcholine (ACh) concentration, locally (by iontophoretic applications) or globally (by stimulating the cholinergic basal forebrain), induces immediate changes in auditory cortical responsiveness [62][63][67••][68][69•]. Pairing of ACh stimulation with a specific auditory stimulus often induces lasting specific changes in responsiveness [70], although probably only when strong enough cholinergic stimulation is applied [69•]. The polarity, magnitude and duration of these changes depend on many experimental parameters. These include the animal's state of arousal ([71•] and see discussion in [62]), the intensity of ACh application [69•][71•][72], the frequency of the basal forebrain stimulation [67••], and, perhaps, even the cortical area (compare [73] with [70]). Furthermore, the polarity of the modifications outlasting the stimulation is not necessarily the same as the polarity during stimulation [69•][70][71•].

The diffuse nature of the cholinergic (and noradrenergic) system makes it a likely candidate for mediating general behavioral control but at the same time questions its sufficiency to generate the observed behavioral specificity. It is possible that specificity is the result of additional mechanisms. On the other hand, the puzzling diversity of cholinergic influences implies that the nature of its effects is far from being understood.

Behavioral expression of physiological modifications

Recanzone et al. [4••] found that expansion of cortical representation of the range of frequencies presented during training on frequency discrimination was correlated with behavioral improvement whereas other modifications were not. Edeline et al. [55••] found that physiological shifts of tuning curves are not sufficient to improve behavioral discrimination. It would be interesting to determine under what conditions frequencies that gain extra representation, for example, by deafferentation or by invasive manipulations, gain behavioral benefits. It would be of further interest to test the effect of preventing local cortical changes, possibly using invasive techniques, on behavioral improvement.

Conclusions

Under a variety of experimental conditions there is an expansion and enhancement of representation of experimentally reinforced input parameters in both cortical and subcortical levels. These changes can be explained by unsupervised Hebbian local rules of plasticity. Indeed, recent studies suggest that a subgroup of the Hebbian rules, subject to behavioral control, operates in the auditory cortex. However, the relation between the cellular expression and intercellular rules of plasticity has not yet been tested directly. Furthermore, the behavioral consequences of these cellular modifications have not been characterized.

Improving behavioral performance probably requires additional, supervised mechanisms, possibly focused at cortical levels. Some aspects of general behavioral control may be mediated by the cholinergic and/or noradrenergic systems. The embodiment of the specific behavioral control, including supervision and its relation to the cholinergic and noradrenergic systems, are at present far from being understood.

Acknowledgements

We would like to thank A Aertsen, M Segal, S Seung for helpful and illuminating remarks and D Shulz for extensive and invaluable discussions on the manuscript. E Ahissar is supported by the Alon Fellowship.

References and recommended reading

1. • Edeline JM , Pham P , Weinberger NM :
   Rapid Development of Learning-Induced Receptive Field Plasticity in the Auditory Cortex.
   Behav Neurosci 1993, 107: 539–551. MEDLINE Cited by
   Shows that the time course of behavioral acquisition of decreased heart rate to conditioned stimulus is correlated with the time course of receptive field modulations: five pairing trials are sufficient.
   

   Return to citation reference [1]

   
   
2. • Weinberger NM , Javid R , Lepan B :
   Long-Term Retention of Learning-Induced Receptive-Field Plasticity in the Auditory Cortex.
   Proc Natl Acad Sci USA 1993, 90: 2394– 2398. MEDLINE Cited by
   Demonstrates that the expanded representation of the conditioned stimulus frequency, following a single session of behavioral aversive conditioning, is retained for at least 8 weeks.
   

   Return to citation reference [1]

   
   
3. Gonzalez-Lima F : Brain Imaging of Auditory Learning Functions in Rats: Studies with Fluorodeoxyglucose Autoradiography and Cytochrome Oxidase Histochemistry. In Advances in Metabolic Mapping Techniques for Brain Imaging of Behavioral and Learning Functions. Edited by Gonzalez-Lima F, Finkenstadt T, Scheich H. Dordrecht: Kluwer Academic Publishers, 1992, 39–109.

   Return to citation reference [1] [2] [3]

   
   
4. •• Recanzone GH , Schreiner CE , Merzenich MM :
   Plasticity in the Frequency Representation of Primary Auditory Cortex Following Discrimination Training in Adult Owl Monkeys.
   J Neurosci 1993, 13: 87–103. MEDLINE Cited by
   A study of the modulations in representation area and average receptive field properties induced by extensive training on difficult frequency discrimination. Shows that passive auditory stimulation while performing a tactual task is not sufficient to expand representation of the trained stimuli, but somewhat sharpens tuning curves and delays minimal response latency. Only expanded representation, however, was correlated with behavioral improvement. Re-mapping was tested for the range of trained frequencies as a whole and not in particular for the behaviorally acquired resolution.
   

   Return to citation reference [1] [2] [3] [4] [5] [6] [7]

   
   
5. Robertson D , Irvine DR :
   Plasticity of Frequency Organization in Auditory Cortex of Guinea Pigs with Partial Unilateral Deafness.
   J Comp Neurol 1989, 282: 456–471. MEDLINE Cited by

   Return to citation reference [1]

   
   
6. Willott JF , Aitkin LM , McFadden SL :
   Plasticity of Auditory Cortex Associated with Sensorineural Hearing Loss in Adult C57BL/6J Mice.
   J Comp Neurol 1993, 329: 402–411. MEDLINE Cited by

   Return to citation reference [1]

   
   
7. • Calford MB , Rajan R , Irvine DR :
   Rapid Changes in the Frequency Tuning of Neurons in Cat Auditory Cortex Resulting from Pure-Tone-Induced Temporary Threshold Shift.
   Neuroscience 1993, 55: 953–964. MEDLINE Cited by
   Application of an intense pure tone was applied to induce temporary threshold shift in cochlear sensitivity. The resulting changes in response areas of neurons in the auditory cortex were complex and involved both expansion and contraction. Thresholds at the expanded frequencies remained high.
   

   Return to citation reference [1]

   
   
8. • Scheich H , Simonis C , Ohl F , Tillein J , Thomas H :
   Functional Organization and Learning-Related Plasticity in Auditory Cortex of the Mongolian Gerbil.
   Prog Brain Res 1993, 97: 135–143. MEDLINE Cited by
   A review of the functional organization of the auditory cortex and its pairing-induced modulations, studied both on a global scale with 2-deoxyglucose, and at the cellular level with electrophysiological recordings. Single and multi-units responses to frequencies adjacent to the conditioned stimulus, but not to the conditioned stimulus frequency, were increased.
   

   Return to citation reference [1] [2]

   
   
9. Hebb DO : The Organization of Behavior. A Neuropsychological Theory, edn 1. New York: Wiley, 1949,

   Return to citation reference [1]

   
   
10. Stent GS :
    A Physiological Mechanism for Hebb's Postulate of Learning.
    Proc Natl Acad Sci USA 1973, 70: 997–1001. MEDLINE Cited by

    Return to citation reference [1]

    
    
11. Stryker MP , Chapman B , Miller KD , Zahs KR :
    Experimental and Theoretical Studies of the Organization of Afferents to Single Orientation Columns in Visual Cortex.
    Cold Spring Harb Symp Quant Biol 1990, 55: 515–527. MEDLINE Cited by

    Return to citation reference [1]

    
    
12. • Singer W :
    Neuronal Representations, Assemblies and Temporal Coherence.
    Prog Brain Res 1993, 95: 461– 474. MEDLINE Cited by
    A review of the possible roles of correlated activity in perceptual processing and in synaptic plasticity. In particular, discusses the significance of non-Hebbian depressions found in visual cortical circuits.
    

    Return to citation reference [1] [2]

    
    
13. • Fregnac Y , Shulz D : Models of Synaptic Plasticity and Cellular Analogs of Learning in the Developing and Adult Visual Cortex. In Advances in Neural and Behavioral Development. Edited by Casagrande VA, Shinkman PG. Norwood, New Jersey: Ablex Publ Corp, 1994, 4: 149–235.
    An extensive review on theoretical and experimental aspects of the role of co-activity in synaptic plasticity. Predictions of the covariance local learning rule are confronted with experimental data measured at the functional and cellular levels.
    

    Return to citation reference [1]

    
    
14. Baranyi A , Feher O :
    Synaptic Facilitation Requires Paired Activation of Convergent Pathways in the Neocortex.
    Nature 1981, 290: 413–415. MEDLINE Cited by

    Return to citation reference [1]

    
    
15. Fregnac Y , Shulz D , Thorpe S , Bienenstock E :
    A Cellular Analogue of Visual Cortical Plasticity.
    Nature 1988, 333: 367– 370. MEDLINE Cited by

    Return to citation reference [1]

    
    
16. Greuel JM , Luhmann HJ , Singer W :
    Pharmacological Induction of Use-Dependent Receptive Field Modifications in the Visual Cortex.
    Science 1988, 242: 74–77. MEDLINE Cited by

    Return to citation reference [1]

    
    
17. Baranyi A , Szente MB , Woody CD :
    Properties of Associative Long-Lasting Potentiation Induced by Cellular Conditioning in the Motor Cortex of Conscious Cats.
    Neuroscience 1991, 42: 321–334. MEDLINE Cited by

    Return to citation reference [1]

    
    
18. Fregnac Y , Shulz D , Thorpe S , Bienenstock E :
    Cellular Analogs of Visual Cortical Epigenesis. I. Plasticity of Orientation Selectivity.
    J Neurosci 1992, 12: 1280–1300. MEDLINE Cited by

    Return to citation reference [1]

    
    
19. Shulz D , Fregnac Y :
    Cellular Analogs of Visual Cortical Epigenesis. II. Plasticity of Binocular Integration.
    J Neurosci 1992, 12: 1301–1318. MEDLINE Cited by

    Return to citation reference [1]

    
    
20. Shulz D , Debanne D , Fregnac Y :
    Cortical Convergence of ON- and OFF-Pathways and Functional Adaptation of Receptive Field Organization in Cat Area 17.
    Prog Brain Res 1993, 95: 191–205. MEDLINE Cited by

    Return to citation reference [1]

    
    
21. Aertsen A , Erb M , Palm G :
    Dynamics of Functional Coupling in the Cerebral Cortex: An Attempt at a Model-Based Interpretation.
    Physica-D 1994, in press.

    Return to citation reference [1]

    
    
22. Sejnowski TJ :
    Storing Covariance with Nonlinearly Interacting Neurons.
    J Math Biol 1977, 4: 303–321. MEDLINE Cited by

    Return to citation reference [1] [2]

    
    
23. EggermonT JJ : The Correlative Brain. Theory and Experiment in Neural Interaction. Berlin: Springer-Verlag, 1990,

    Return to citation reference [1] [2]

    
    
24. Uttley AM : Information Transmission in the Nervous System. London: Academic Press, 1976,

    Return to citation reference [1]

    
    
25. Levy WB : A Computational Approach to Hippocampal Function. In Computational Models of Learning in Simple Neural Systems. Edited by Hawkins RD, Bower GH. New York: Academic Press, 1989, 243–305.

    Return to citation reference [1]

    
    
26. Aertsen A , Gerstein GL , Habib MK , Palm G :
    Dynamics of Neuronal Firing Correlation: Modulation of ''Effective Connectivity''.
    J Neurophysiol 1989, 61: 900–917. MEDLINE Cited by

    Return to citation reference [1]

    
    
27. Ahissar E , Vaadia E , Ahissar M , Bergman H , Arieli A , Abeles M :
    Dependence of Cortical Plasticity on Correlated Activity of Single Neurons and on Behavioral Context.
    Science 1992, 257: 1412–1415. MEDLINE Cited by

    Return to citation reference [1] [2] [3] [4] [5] [6]

    
    
28. Abeles M : Local Cortical Circuits: an Electrophysiological Study. Berlin: Springer, 1982,

    Return to citation reference [1]

    
    
29. Kruger J (Ed) : Neuronal cooperativity. Berlin: Springer-Verlag, 1991,

    Return to citation reference [1]

    
    
30. Sakurai Y :
    Dependence of Functional Synaptic Connections of Hippocampal and Neocortical Neurons on Types of Memory.
    Neurosci Lett 1993, 158: 181–184. MEDLINE Cited by

    Return to citation reference [1]

    
    
31. Fregnac Y , Shulz D : Hebbian Synapses in the Visual Cortex. In Seeing Contour and Colour. Edited by Kulikowski KK. Oxford: Pergamon Press, 1989, 711–718.

    Return to citation reference [1]

    
    
32. Di Prisco VG :
    Hebb Synaptic Plasticity.
    Prog Neurobiol 1984, 22: 89–102. MEDLINE Cited by

    Return to citation reference [1]

    
    
33. Sejnowski TJ , Tesauro GJ : The Hebb Rule for Synaptic Plasticity: Implementations and Applications. In Neural Models of Plasticity. Edited by Byrne JH, Berry WO. San Diego: Academic Press, 1989, 94–103.

    Return to citation reference [1]

    
    
34. Brown TH , Kairiss EW , Keenan CL :
    Hebbian Synapses: Biophysical Mechanisms and Algorithms.
    Annu Rev Neurosci 1990, 13: 475–511. MEDLINE Cited by

    Return to citation reference [1] [2] [3]

    
    
35. Sutton RS , Barto AG :
    Toward a Modern Theory of Adaptive Networks: Expectation and Prediction.
    Psychol Rev 1981, 88: 135– 170. MEDLINE Cited by

    Return to citation reference [1] [2] [3]

    
    
36. Donegan NH , Gluck MA , Thompson RF : Integrative Behavioral and Biological Models of Classical Conditioning. In The Psychology of Learning and Motivation. Edited by Bower GH. New York: Academic Press, 1989, 23: 109–156.

    Return to citation reference [1]

    
    
37. Tesauro G :
    Simple Neural Models of Classical Conditioning.
    Biol Cybern 1986, 55: 187–200. MEDLINE Cited by

    Return to citation reference [1]

    
    
38. Ahissar E : Examination of Models for Learning in Behaving Monkey's Cortex [PhD Thesis]. Jerusalem: The Hebrew University, 1991,

    Return to citation reference [1] [2]

    
    
39. Barlow HB :
    Unsupervised Learning.
    Neural Computation 1989, 1: 295–311. Cited by

    Return to citation reference [1] [2]

    
    
40. Bienenstock E , Cooper LN , Munro PW :
    Theory for the Development of Neuron Selectivity: Orientation Specificity and Binocular Interaction in Visual Cortex.
    J Neurosci 1982, 2: 32– 48. MEDLINE Cited by

    Return to citation reference [1]

    
    
41. Farley J , Alkon DL :
    Cellular Mechanisms of Learning, Memory, and Information Storage.
    Annu Rev Psychol 1985, 36: 419– 494. MEDLINE Cited by

    Return to citation reference [1]

    
    
42. • Shu ZJ , Swindale NV , Cynader MS :
    Spectral Motion Produces an Auditory After-Effect.
    Nature 1993, 364: 721– 723. MEDLINE Cited by
    Adaptation to a continuous monotonic change of the centre frequency of a broad-band noise for a few minutes yielded a consistent judgement bias, towards the opposite direction, for subsequent similar stimuli. This after-effect was transferred between ears but not between processing channels (probed by superimposed spectral peak versus a superimposed spectral notch) and depended on adaption velocity.
    

    Return to citation reference [1]

    
    
43. • Kapadia MK , Gilbert CD , Westheimer G :
    A Quantitative Measure for Short-Term Cortical Plasticity in Human Vision.
    J Neurosci 1994, 14: 451–457. MEDLINE Cited by
    A psychophysical study measuring the effect of an artificial scotoma, produced by a mask, on localization. A rapidly induced bias in position estimation towards the interior of the scotoma was found. It was interpreted as deriving from expansion of receptive fields within the scotoma to neighboring areas without an accompanying update of the readout mechanism.
    

    Return to citation reference [1]

    
    
44. McMullen NT , De Venecia AK :
    Thalamocortical Patches in Auditory Neocortex.
    Brain Res 1993, 620: 317–322. MEDLINE Cited by

    Return to citation reference [1]

    
    
45. • Clarke S , De Ribaupierre F , Rouiller EM , De Ribaupierre Y :
    Several Neuronal and Axonal Types Form Long Intrinsic Connections in the Cat Primary Auditory Cortical Field (AI).
    Anat Embryol (Berl) 1993, 188: 117–138. MEDLINE Cited by
    Describes the topological organization, and the types of neurons and axons involved in the local, patchy and anisotropic connectivity in the primary auditory cortical field. Connectivity is slightly biased in favor of connections (up to 7 mm in length) from areas of higher best-frequencies to areas of lower best-frequencies.
    

    Return to citation reference [1]

    
    
46. Morel A , Garraghty PE , Kaas JH :
    Tonotopic Organization, Architectonic Fields, and Connections of Auditory Cortex in Macaque Monkeys.
    J Comp Neurol 1993, 335: 437–459. MEDLINE Cited by

    Return to citation reference [1]

    
    
47. • Woody CD , Wang X , Gruen E :
    Response to Acoustic Stimuli Increases in the Ventral Cochlear Nucleus After Stimulus Pairing.
    Neuroreport 1994, 5: 513–515. MEDLINE Cited by
    The responsiveness of neurons of the ventral cochlear nucleus was modified after either backward (an unconditioned stimulus, glabella tap and hypothalamic electrical stimulation, precedes the conditioned stimulus, a click) or forward conditioning (an unconditioned stimulus follows the conditioned stimulus). Modulations were attributed to increased attentiveness. Time course of modifications was not described.
    

    Return to citation reference [1]

    
    
48. • McIntosh AR , Gonzalez-Lima F :
    Network Analysis of Functional Auditory Pathways Mapped with Fluorodeoxyglucose: Associative Effects of a Tone Conditioned as a Pavlovian Excitor or Inhibitor.
    Brain Res 1993, 627: 129–140. MEDLINE Cited by
    Two groups of rats were trained with a conditioned stimulus associated as either an excitor (included in reinforced trials) or as an inhibitor, respectively. Differential effects, measured with flurodeoxyglucose, indicate that multiple levels within the auditory system are subject to changes that depend on the associative significance of the stimulus.
    

    Return to citation reference [1] [2]

    
    
49. Oleson TD , Ashe JH , Weinberger NM :
    Modifications of Auditory and Somatosensory System Activity During Pupillary Conditioning in the Paralyzed Cat.
    J Neurophysiol 38: 1114–1139. MEDLINE Cited by

    Return to citation reference [1]

    
    
50. • Rajan R , Irvine DRF , Wise LZ , Heil P :
    Effect of Unilateral Partial Cochlear Lesions in Adult Cats on the Representation of Lesioned and Unlesioned Cochleas in Primary Auditory-Cortex.
    J Comp Neurol 1993, 338: 17–49. MEDLINE Cited by
    The effect of a restricted unilateral cochlear lesion was measured after 2–11 months. Ipsilaterally, there was no change. Contralaterally, the representation of frequencies at the edge of the lesion expanded by up to ~ 2.6 mm and neurons re-gained 'normal' receptive field properties at their new characteristic frequencies. Characteristics of reorganization point to a locus before binaural convergence.
    

    Return to citation reference [1] [2]

    
    
51. Pons TP , Garraghty PE , Ommaya AK , Kaas JH , Taub E , Mishkin M :
    Massive Cortical Reorganization After Sensory Deafferentation in Adult Macaques.
    Science 1991, 252: 1857–1860. MEDLINE Cited by

    Return to citation reference [1]

    
    
52. Merzenich MM , Kaas JH , Wall JT , Sur M , Nelson RJ , Felleman DJ :
    Progression of Change Following Median Nerve Section in the Cortical Representation of the Hand in Areas 3b and 1 in Adult Owl and Squirrel Monkeys.
    Neuroscience 1983, 10: 639–665. MEDLINE Cited by

    Return to citation reference [1]

    
    
53. Garraghty PE , Kaas JH :
    Dynamic Features of Sensory and Motor Maps.
    Curr Opin Neurobiol 1992, 2: 522–527. MEDLINE Cited by

    Return to citation reference [1]

    
    
54. • Weinberger NM :
    Dynamic Regulation of Receptive Fields and Maps in the Adult Sensory Cortex.
    Annu Rev Neurosci 1995, in press.
    A comprehensive review of the various methodologies and main findings of plasticity in sensory cortices. Accumulated data is analyzed in view of answered and currently open, basic questions.
    

    Return to citation reference [1] [2]

    
    
55. •• Edeline JM , Weinberger NM :
    Receptive Field Plasticity in the Auditory Cortex During Frequency Discrimination Training: Selective Retuning Independent of Task Difficulty.
    Behav Neurosci 1993, 107: 82–103. MEDLINE Cited by
    Similar modulations are found in receptive field properties following both difficult and easy differential conditioning, although only in the easy case it is accompanied by a discriminative response of heart rate.
    

    Return to citation reference [1] [2] [3] [4]

    
    
56. Dudai Y :
    Why 'Learning' and 'Memory' Should be Redefined (or, an Agenda for Focused Reductionism).
    Concepts Neurosci 1992, 3: 99–121.

    Return to citation reference [1]

    
    
57. Recanzone GH , Merzenich MM , Schreiner CE :
    Changes in the Distributed Temporal Response Properties of SI Cortical Neurons Reflect Improvements in Performance on a Temporally Based Tactile Discrimination Task.
    J Neurophysiol 1992, 67: 1071–1091. MEDLINE Cited by

    Return to citation reference [1]

    
    
58. • Seung HS , Sompolinski H :
    Simple Models for Reading Neuronal Population Codes.
    Proc Natl Acad Sci USA 1993, 90: 10749–10753. MEDLINE Cited by
    An analysis of a linear network that improves discrimination by optimizing the 'readout' of a population vector. Discusses receptive field characteristics that yield maximal information for a given task.
    

    Return to citation reference [1]

    
    
59. • Ahissar M , Hochstein S :
    Attentional Control of Early Perceptual Learning.
    Proc Natl Acad Sci USA 1993, 90: 5718–5722. MEDLINE Cited by
    A psychophysical study of humans learning simple visual tasks. Demonstrates that intensive, spatially attentive training with a stimulus does not yield substantial improvement without task- specific selective attention.
    

    Return to citation reference [1]

    
    
60. Weinberger NM :
    Learning-Induced Changes of Auditory Receptive Fields.
    Curr Opin Neurobiol 1993, 3: 570–577. MEDLINE Cited by

    Return to citation reference [1]

    
    
61. Ahissar M , Ahissar E , Bergman H , Vaadia E :
    Encoding of Sound Source and Movement: the Activity of Single Neurons and Interactions Between Adjacent Neurons in the Primary Auditory Cortex of Monkeys.
    J Neurophysiol 1992, 67: 203–215. MEDLINE Cited by

    Return to citation reference [1]

    
    
62. Hars B , Maho C , Edeline JM , Hennevin E :
    Basal Forebrain Stimulation Facilitates Tone-Evoked Responses in the Auditory Cortex of Awake Rat.
    Neuroscience 1993, 56: 61–74. MEDLINE Cited by

    Return to citation reference [1] [2] [3]

    
    
63. McKenna TM , Ashe JH , Hui GK , Weinberger NM :
    Muscarinic Agonists Modulate Spontaneous and Evoked Unit Discharge in Auditory Cortex of Cat.
    Synapse 1988, 2: 54–68. MEDLINE Cited by

    Return to citation reference [1] [2]

    
    
64. •• Naatanen R , Schroger E , Karakas S , Tervanieme M , Paavilainen P :
    Development of a Memory Trace for a Complex Sound in the Human Brain.
    Neuroreport 1993, 4: 503–506. MEDLINE Cited by
    An EEG component, probably generated in the auditory cortex, appeared and increased during training on auditory discrimination task. This component did not emerged when the same stimuli were merely presented to non-aware subjects. The same component was recorded from naive subjects that exhibited good performance before training.
    

    Return to citation reference [1]

    
    
65. Alho K , Huotilainen M , Tiitinen H , Ilmoniemi RJ , Knuutila J , Naatanen R :
    Memory-Related Processing of Complex Sound Patterns in Human Auditory Cortex: a MEG Study.
    Neuroreport 1993, 4: 391–394. MEDLINE Cited by

    Return to citation reference [1]

    
    
66. Thorndike EL : Human Nature and the Social Order. New York: Macmillan, 1940,

    Return to citation reference [1]

    
    
67. •• Metherate R , Ashe JH :
    Nucleus Basalis Stimulation Facilitates Thalamocortical Synaptic Transmission in the Rat Auditory Cortex.
    Synapse 1993, 14: 132– 143. MEDLINE Cited by
    The effects of brief tetanic stimulations of the nucleus basalis, at intensities that often activate local cortical EEG, were measured by field potentials, single neuronal discharges and intracellularly recorded postsynaptic potentials. The stimulations increased amplitude and reduced latencies of auditory thalamocortical transmission for several seconds. The magnitude of facilitation depended on stimulation frequency and was reduced by cortically applied atropine.
    

    Return to citation reference [1] [2]

    
    
68. McKenna TM , Ashe JH , Weinberger NM :
    Cholinergic Modulation of Frequency Receptive Fields in Auditory Cortex: I. Frequency-Specific Effects of Muscarinic Agonists.
    Synapse 1989, 4: 30–43. MEDLINE Cited by

    Return to citation reference [1]

    
    
69. • Edeline JM , Hars B , Maho C , Hennevin E :
    Transient and Prolonged Facilitation of Tone-Evoked Responses Induced by Basal Forebrain Stimulation in the Rat Auditory Cortex.
    Exp Brain Res 1994, 97: 373–386. MEDLINE Cited by
    In anesthetized rats, pairing a tone with strong (i.e. yielding global EEG desynchronization) stimulations of the cholinergic basal forebrain resulted, on the average, in lasting (tested up to 15 min) facilitation of the tone-evoked auditory cortical responses. Similar pairing with weaker (non-desynchronizing) stimulations resulted, on the average, in transient changes which did not outlast the pairing period. The polarity and magnitude of modifications, during and after stimulations, were notably variable. Facilitations were specific to the stimulated hemisphere and were prevented by systemic atropine application.
    

    Return to citation reference [1] [2] [3] [4]

    
    
70. Metherate R , Weinberger NM :
    Cholinergic Modulation of Responses to Single Tones Produces Tone-Specific Receptive Field Alterations in Cat Auditory Cortex.
    Synapse 1990, 6: 133–145. MEDLINE Cited by

    Return to citation reference [1] [2] [3]

    
    
71. • Edeline JM , Maho C , Hars B , Hennevin E :
    Non-Awaking Basal Forebrain Stimulation Enhances Auditory Cortex Responsiveness During Slow-Wave Sleep.
    Brain Res 1994, 636: 333– 337. MEDLINE Cited by
    A demonstration of the considerable variability of the effects of cholinergic stimulations. During slow-wave sleep, tone-evoked cortical responses were facilitated following pairing of a tone with weak (non-awaking) basal forebrain stimulations, but not following pairing with strong (awaking) stimulations. During wakefulness, pairing with strong stimulations yielded atropine- blocked post-pairing facilitations but atropine-enhanced during- pairing depressions.
    

    Return to citation reference [1] [2] [3]

    
    
72. Metherate R , Ashe JH :
    Basal Forebrain Stimulation Modifies Auditory Cortex Responsiveness by an Action at Muscarinic Receptors.
    Brain Res 1991, 559: 163–167. MEDLINE Cited by

    Return to citation reference [1]

    
    
73. Metherate R , Tremblay N , Dykes RW :
    Acetylcholine Permits Long-Term Enhancement of Neuronal Responsiveness in Cat Primary Somatosensory Cortex.
    Neuroscience 1987, 22: 75–81. MEDLINE Cited by

    Return to citation reference [1]

    
    
74. Artola A , Brocher S , Singer W :
    Differential Voltage- Dependent Threshold for Inducing Long-Term Depression and Long- Term Potentiation in Slices of Rat Visual Cortex.
    Nature 1990, 347: 69–72. MEDLINE Cited by

    Return to citation reference [1]

    
    
75. Lisman J :
    A Mechanism for the Hebb and the Anti-Hebb Processes Underlying Learning and Memory.
    Proc Natl Acad Sci USA 1989, 86: 9574–9578. MEDLINE Cited by

    Return to citation reference [1]

    
    

Author Contacts

Return to author list

Copyright