Declarative memory consolidation: mechanisms acting during human sleep

Of late, an increasing number of studies have shown a strong relationship between sleep and memory. Here we summarize a series of our own studies in humans supporting a beneficial influence of slow-wave sleep (SWS) on declarative memory formation, and try to identify some mechanisms that might underlie this influence. Specifically, these experiments show that declarative memory benefits mainly from sleep periods dominated by SWS, whereas there is no consistent benefit of this memory from periods rich in rapid eye movement (REM) sleep. A main mechanism of declarative memory formation is believed to be the reactivation of newly acquired memory representations in hippocampal networks that stimulates a transfer and integration of these representations into neocortical neuronal networks. Consistent with this model, spindle activity and slow oscillation-related EEG coherence increase during early sleep after intense declarative learning in humans, signs that together point toward a neocortical reprocessing of the learned material. In addition, sleep seems to provide an optimal milieu for declarative memory reprocessing and consolidation by reducing cholinergic activation and the cortisol feedback to the hippocampus during SWS.     

Impaired memory consolidation in rats produced with beta-adrenergic blockade

Despite abundant evidence that systemic administration of adrenergic drugs and hormones can produce retrograde memory enhancement, the literature contains no clear demonstration that postlearning systemic administration of adrenergic antagonists produces retrograde amnesia. Here we demonstrate retrograde amnesia for a stressful learning task (a spatial water maze) with systemic administration of the beta-adrenergic antagonist propranolol (5 mg/kg). The amnesic effect of the drug depended on the degree of learning in the subjects: Propranolol caused a robust retrograde amnesia in "good learners," but did not significantly affect memory in "poor learners." The findings provide critical additional support for the hypothesis that postlearning adrenergic activation modulates memory consolidation processes after emotionally stressful events and help explain previous failures to detect memory impairment after systemic administration of adrenergic blocking drugs.     

Cerebral asymmetries in sleep-dependent processes of memory consolidation

Preference for previously seen, unfamiliar objects reflects a memory bias on affective judgment, known as the “mere exposure effect” (MEE). Here, we investigated the effect of time, post-exposure sleep, and the brain hemisphere solicited on preference generalization toward objects viewed in different perspectives. When presented in the right visual field (RVF), which promotes preferential processing in the left hemisphere, same and mirrored exemplars were preferred immediately after exposure. MEE generalized to much dissimilar views after three nights of sleep. Conversely, object presentation in the left visual field (LVF), promoting right hemisphere processing, elicited a MEE for same views immediately after exposure, then for mirror views after sleep. Most importantly, sleep deprivation during the first post-exposure night, although followed by two recovery nights, extinguished MEE for all views in the LVF but not in the RVF. Besides demonstrating that post-exposure time and sleep facilitate the generalization process by which we integrate various representations of an object, our results suggest that mostly in the right hemisphere, sleep may be mandatory to consolidate the memory bias underlying affective preference. These interhemispheric differences tentatively call for a reappraisal of the role of cerebral asymmetries in wake- and sleep-dependent processes of memory consolidation.     

System consolidation of memory during sleep

Over the past two decades, research has accumulated compelling evidence that sleep supports the formation of long-term memory. The standard two-stage memory model that has been originally elaborated for declarative memory assumes that new memories are transiently encoded into a temporary store (represented by the hippocampus in the declarative memory system) before they are gradually transferred into a long-term store (mainly represented by the neocortex), or are forgotten. Based on this model, we propose that sleep, as an offline mode of brain processing, serves the ‘active system consolidation’ of memory, i.e. the process in which newly encoded memory representations become redistributed to other neuron networks serving as long-term store. System consolidation takes place during slow-wave sleep (SWS) rather than rapid eye movement (REM) sleep. The concept of active system consolidation during sleep implicates that (a) memories are reactivated during sleep to be consolidated, (b) the consolidation process during sleep is selective inasmuch as it does not enhance every memory, and (c) memories, when transferred to the long-term store undergo qualitative changes. Experimental evidence for these three central implications is provided: It has been shown that reactivation of memories during SWS plays a causal role for consolidation, that sleep and specifically SWS consolidates preferentially memories with relevance for future plans, and that sleep produces qualitative changes in memory representations such that the extraction of explicit and conscious knowledge from implicitly learned materials is facilitated.     

Stimulation of hippocampal adenylyl cyclase activity dissociates memory consolidation processes for response and place learning

Procedural and declarative memory systems are postulated to interact in either a synergistic or a competitive manner, and memory consolidation appears to be a highly critical stage for this process. However, the precise cellular mechanisms subserving these interactions remain unknown. To investigate this issue, 24-h retention performances were examined in mice given post-training intrahippocampal injections of forskolin (FK) aiming at stimulating hippocampal adenylyl cyclases (ACs). The injection was given at different time points over a period of 9 h following acquisition in either an appetitive bar-pressing task or water-maze tasks challenging respectively "response memory" and "place memory." Retention testing (24 h) showed that FK injection altered memory formation only when given within a 3- to 6-h time window after acquisition but yielded opposite memory effects as a function of task demands. Retention of the spatial task was impaired, whereas retention of both the cued-response in the water maze and the rewarded bar-press response were improved. Intrahippocampal injections of FK produced an increase in pCREB immunoreactivity, which was strictly limited to the hippocampus and lasted less than 2 h, suggesting that early effects (0-2 h) of FK-induced cAMP/CREB activation can be distinguished from late effects (3-6 h). These results delineate a consolidation period during which specific cAMP levels in the hippocampus play a crucial role in enhancing memory processes mediated by other brain regions (e.g., dorsal or ventral striatum) while eliminating interference by the formation of hippocampus-dependent memory.     

Delayed working memory consolidation during the attentional blink

After the detection of a target (T1) in a rapid stream of visual stimuli, there is a period of 400–600 msec during which a subsequent target (T2) is missed. This impairment in performance has been labeled the attentional blink. Recent theories propose that the attentional blink reflects a bottleneck in working memory consolidation such that T2 cannot be consolidated until after T1 is consolidated, and T2 is therefore masked by subsequent stimuli if it is presented while T1 is being consolidated. In support of this explanation, Giesbrecht & Di Lollo (1998) found that when T2 is the final item in the stimulus stream, no attentional blink is observed, because there are no subsequent stimuli that might mask T2. To provide a direct test of this explanation of the attentional blink, in the present study we used the P3 component of the event-related potential waveform to track the processing of T2. When T2 was followed by a masking item, we found that the P3 wave was completely suppressed during the attentional blink period, indicating that T2 was not consolidated in working memory. When T2 was the last item in the stimulus stream, however, we found that the P3 wave was delayed but not suppressed, indicating that T2 consolidation was not eliminated but simply delayed. These results are consistent with a fundamental limit on the consolidation of information in working memory.     

Fractionating working memory: consolidation and maintenance are independent processes

In the present study, we required subjects to remember simple objects that were masked to interrupt consolidation and allow us to estimate the rate of information accrual in visual working memory. We compared a consolidation-baseline condition with a consolidation-during-maintenance condition in which subjects needed to remember a set of unmasked items and then were shown to-be-remembered masked items. We hypothesized that if the control processes of consolidation and maintenance are performed by common mechanisms, then consolidation should be less efficient when performed during maintenance than when performed alone. However, we found that an identical amount of information was encoded per unit time in the two conditions. These results indicate that working memory consolidation is not slowed by maintenance and suggest a two-step model of encoding in visual working memory.     

Differential activity profile of cAMP-dependent protein kinase isoforms during long-term memory consolidation in the crab Chasmagnathus

The isoforms of cAMP-dependent protein kinase (PKA) show distinct biochemical properties and subcellular localization, suggesting different physiological functions, and conferring the fine-tuning between the activation of cAMP-PKA cascade and the cellular response. The critical role of PKA in memory and synaptic plasticity has been extensively demonstrated both in vertebrates and invertebrates, but the role of PKA isoforms is a matter of debate. Here we present experimental data showing differential PKA activation profiles after two different experiences: an instance of associative contextual learning (context-signal learning) and a single exposure to a novel context, both in the learning and memory model of the crab Chasmagnathus. Differences were found in the temporal course of activation and in the involvement of PKA isoforms. We found increased PKA activity immediately and 6 h after context-signal training correlating with the critical periods during which pharmacological inhibition of PKA disrupts memory formation. In contrast, PKA activity increased immediately but not 6 h after single exposure to a novel context. The amounts of PKA I and PKA II holoenzymes were analyzed to determine changes in holoenzyme levels and/or differential activation induced by both experiences. Results indicate that context-induced PKA activation is at least in part due to PKA II, and that PKA activation 6 h after context-signal learning coincides with an increase in the total level of PKA I. Considering the higher sensitivity of PKA I to cAMP, its increment can account for the PKA activation found 6 h after training and is proposed as a novel mechanism providing the prolonged PKA activation during memory consolidation.     

Vasopressin modulates the activity of nicotinic cholinergic mechanisms during memory retrieval in mice

Lysine vasopressin (0.03 micrograms/kg, sc) enhanced retention test performance on a one-trial step-through inhibitory avoidance task when injected into male Swiss mice 20 min before the retention test. Tests were done 48 h following training. A low dose of the vasopressin antagonist AAVP (0.01 microgram/kg, sc, 20 min prior to testing) did not significantly affect retention test performance, whereas a higher dose (0.03 microgram/kg, sc) impaired it. Neither lysine vasopressin nor AAVP when given prior to testing modified latencies to step-through of mice that had not received a footshock during training. The simultaneous administration of AAVP (0.01 microgram/kg, sc) prevented the enhancement of retention test performance induced by lysine vasopressin. The influence of lysine vasopressin on retention test performance was antagonized by the simultaneous administration of mecamylamine (5 mg/kg, sc) but not by hexamethonium (5 mg/kg, sc), atropine (0.5 mg/kg, sc), or methylatropine (0.5 mg/kg, sc). A modulatory role of vasopressin on the activity of central cholinergic nicotinic mechanisms which probably operate at the time of testing is suggested.     

Attentional effects on orientation judgements are dependent on memory consolidation processes

Are the effects of memory and attention on perception synergistic, antagonistic, or independent? Tested separately, memory and attention have been shown to affect the accuracy of orientation judgements. When multiple stimuli are presented sequentially versus simultaneously, error variance is reduced. When a target is validly cued, precision is increased. What if they are manipulated together? We combined memory and attention manipulations in an orientation judgement task to answer this question. Two circular gratings were presented sequentially or simultaneously. On some trials a brief luminance cue preceded the stimuli. Participants were cued to report the orientation of one of the two gratings by rotating a response grating. We replicated the finding that error variance is reduced on sequential trials. Critically, we found interacting effects of memory and attention. Valid cueing reduced the median, absolute error only when two stimuli appeared together and improved it to the level of performance on uncued sequential trials, whereas invalid cueing always increased error. This effect was not mediated by cue predictiveness; however, predictive cues reduced the standard deviation of the error distribution, whereas nonpredictive cues reduced “guessing”. Our results suggest that, when the demand on memory is greater than a single stimulus, attention is a bottom-up process that prioritizes stimuli for consolidation. Thus attention and memory are synergistic.     

Differential role of the hippocampal endocannabinoid system in the memory consolidation and retrieval mechanisms

CB1 cannabinoid receptors are abundantly expressed in the brain, with large concentrations present in the hippocampus, a brain structure essential for memory processing. In the present study, we have investigated the possible modulatory role of the endocannabinoid system in the dorsal hippocampus upon the different phases of memory processing of an aversive task. AM251, a selective antagonist of CB1 receptors, and anandamide, an endogenous agonist of cannabinoid receptors, were bilaterally infused into the dorsal hippocampus of male Wistar rats either before training, immediately after training, or before test in the step-down inhibitory avoidance (IA) task. Results showed that pre-training infusion of CB1 drugs did not influence the acquisition of the task. In contrast, post-training infusion of the CB1 antagonist disrupted while the antagonist facilitated memory consolidation of IA. The post-training results demonstrate that memory consolidation depends on the integrity of the endocannabinoid system in the CA1 region of the dorsal hippocampus. While we still have no direct proof of endocannabinoids released there after an aversive task such as IA, these results suggests that (a) AM251 acts blocking the binding of endogenously released cannabinoids and (b) exogenously supplemented anandamide may be adding its contribution to the action of the endogenously released pool. Considering our data and the higher density of CB1 receptors present in the GABAergic interneurons, we propose them as the putative target of the endocannabinoid modulation of memory, a hypothesis that needs to be proven. In addition, pre-test infusion of the CB1 receptor antagonist facilitated while infusion of the agonist did not affect memory retrieval of IA. The completely opposite action of the same drug upon memory at the post-training (consolidation) and pre-test (recall) contexts suggests that some durable change took place in the CA1 region during the consolidation process that modified the logical attributes of the pharmacological response, i.e., the drug response changed from memory disruption to memory facilitation. A similar phenomenon was previously described by us in the M4 cholinergic muscarinic subsystem in the hippocampus for the same task (Diehl, F., Fürstenau, L. O., Sanchez, G., Camboim, C., de Oliveira Alvares, L., Lanziotti, V. B., et al. (2007). Facilitatory effect of the intra-hippocampal pretest administration of MT3 in the inhibitory avoidance task. Behavioral Brain Research, 177(2), 227-231), but the biological nature of such change in the local neural circuitry remains to be investigated.     

Amygdala modulation of memory consolidation: interaction with other brain systems

There is a strong consensus that the amygdala is involved in mediating influences of emotional arousal and stress on learning and memory. There is extensive evidence that the basolateral amygdala (BLA) is a critical locus of integration of neuromodulatory influences regulating the consolidation of several forms of memory. Many drug and stress hormone influences converge in activating the release of norepinephrine (NE) within the BLA. Evidence from studies using in vivo microdialysis and high-performance liquid chromatography indicates that increases in amygdala NE levels assessed following inhibitory avoidance training correlate highly with subsequent retention. Other evidence indicates that NE influences on memory consolidation require muscarinic cholinergic activation within the BLA provided by projections from the nucleus basalis magnocellularis (NB). Evidence from several experiments indicates that activation of the BLA plays an essential role in modulating memory consolidation processes involving other brain regions. These findings provide strong support for the hypothesis that the BLA plays a critical role in regulating the consolidation of lasting memories of significant experiences.     

Commentary: cognitive-affective mechanisms and processes in autobiographical memory

This commentary highlights some of the interesting points to emerge from the preceding papers about the self, social, and directive functions of autobiographical memory. Additionally some cognitive functions are also considered and especially the way in which autobiographical memory supports, constrains, and maintains the goals of the self. Directions for future research into the self, social, directive, and cognitive-affective functions and processes of autobiographical memory are reviewed. Emphasis is placed on future research into the function of autobiographical memory in representations of attachment.     

Contextual modulation of memory consolidation

We investigate olfactory memory consolidation in honeybees. Three experiments are reported that include 1024 animals in 28 experimental groups. After one pairing of odorant and sucrose reward, retention is typically nonmonotonic with a minimum 3 min after conditioning. This corresponds to the “Kamin effect” in vertebrates; the postminimum rise in retention is usually interpreted as reflecting memory consolidation. First, we test for the generality of this effect across four different odorants. The postminimum rise in retention was reproducibly observed for 1-hexanol but not for 1-octanol, limonene, or geraniol. Second, we investigate whether previous learning about the training context modulates subsequent memory consolidation. On the day before training, a reward was applied either upon placement into the future training context for 1 min, halfway during exposure or just before removal from the context. In the latter group, the 3-min minimum in retention was eliminated; thus, in that group, forward pairings of context and reward (i.e., context exposure begins before reward is applied) lead to an associative context memory that can modulate subsequent olfactory memory consolidation. Third, we found no evidence for a modulation of olfactory memory consolidation by pre-exposure to the odorant.     

Myosin II motor activity in the lateral amygdala is required for fear memory consolidation

Learning induces dynamic changes to the actin cytoskeleton that are required to support memory formation. However, the molecular mechanisms that mediate filamentous actin (F-actin) dynamics during learning and memory are poorly understood. Myosin II motors are highly expressed in actin-rich growth structures including dendritic spines, and we have recently shown that these molecular machines mobilize F-actin in response to synaptic stimulation and learning in the hippocampus. In this study, we report that Myosin II motors in the rat lateral amygdala (LA) are essential for fear memory formation. Pretraining infusions of the Myosin II inhibitor, blebbistatin (blebb), disrupted long term memory, while short term memory was unaffected. Interestingly, both post-training and pretesting infusions had no effect on memory formation, indicating that Myosin II motors operate during or shortly after learning to promote memory consolidation. These data support the idea that Myosin II motor-force generation is a general mechanism that supports memory consolidation in the mammalian CNS.