Anterior Thalamic Nuclei: A Critical Substrate For Non-spatial Paired-associate Memory in Rats Part 3
Dec 20, 2023
3.2 | Spatial working memory in the RAM
As expected, both groups of ATN-lesion rats showed severely impaired performance when spatial working memory was tested in the eight-arm RAM (Figure 3). The initial similarity in performance across groups is because most rats ran for 10 min but made relatively few arm entries at the start of testing.
Working memory and memory are two very important memory methods in the human brain. Although there are some differences between these two memory methods, they are both equally helpful for people's performance in study, work, life, and other aspects.
First of all, working memory means that we temporarily remember the information or tasks we need to use in a short period. In other words, it is the memory capacity we use when processing new things. Memory refers to the information or experiences we retain over a long period. This information can help us understand and solve problems in future use.
Although these two types of memory have different functions, they are not separate from each other. Working memory and memory influence each other. During working memory, we need to temporarily store and process new information in a short period and connect it with previous knowledge and experience. This kind of connection and reference is what memory uses.
In addition, good working memory can also be improved through training. Research shows that the connection between memory and working memory goes both ways. By constantly exercising and training working memory, we can improve the brain's memory capacity and strengthen our memory. Therefore, as long as we keep actively exercising our brains and strengthening memory training, we can effectively improve our working memory and memory.
In daily life, we need to develop good habits and actively use working memory and memory to help us complete tasks better. For example, we can strengthen our memory of new knowledge through continuous refining and review. Only by continuously improving our working memory and memory can we be more comfortable in realizing our potential in study, work, and life. It can be seen that we need to improve memory, and Cistanche deserticola can significantly improve memory because Cistanche deserticola is a traditional Chinese medicinal material that has many unique effects, one of which is to improve memory. The efficacy of minced meat comes from the various active ingredients it contains, including acid, polysaccharides, flavonoids, etc. These ingredients can promote brain health in various ways.
Click know 10 ways to improve memory
The number of errors made by Sham- and ATN-lesion rats diverged on Day 2 of training after which both ATN-lesion groups showed no evidence of improvement compared to both Sham groups (Group, F3,27 = 39.66, p < 0.001; Group Day, F27,243 = 4.75, p < 0.001). The two sham groups did not differ, but they both differed markedly (p < 0.001) from each of the ATN-lesion groups, which did not differ. Rats in both ATN-lesion groups also made fewer correct arm choices before the first error, whereas rats in both Sham groups progressively entered more arms before making an error as testing continued (Group, F3,27 = 27.53, p < 0.001; each Sham group vs. each ATN group, p < 0.001; Group Day, F27,243 = 2.21, p < 0.001).
Impaired spatial working memory can be regarded as a benchmark measure of ATN-lesion effects (Aggleton & Nelson, 2015). As the two ATN-lesion groups showed a similar level of impairment in spatial working memory, subsequent comparisons on the non-spatial tasks are unlikely to be influenced by lesion differences between these two groups.
3.3 | Non-spatial simple discrimination tasks
Acquisition of both the simple odor and simple object, go-no-go discrimination tasks are shown in Figure 4a,b. All rats rapidly acquired these tasks. The latency difference scores for the four groups did not differ in the simple odor discrimination task (Group, F3,27 = 0.97, p = 0.42; Group Day, F15,135 = 1.4, p = 0.15). The groups took an average of 4–5 days to reach the criterion in the simple odor discrimination task (Group, F3,27 = 1.99, p = 0.13).
The four groups also learned the simple object discrimination task without any significant differences (Group, F3,27 = 1.07, p = 0.37; Group Day, F18,162 = 0.97, p = 0.49). Rats took an average of 5–6 days to reach the criterion on the simple object discrimination task (Group, F3,27 = 0.30, p = 0.82).
3.4 | Non-spatial paired-associate tasks
3.4.1 | Acquisition
The mean running latency during acquisition by the four groups is shown in Figure 5. Latencies during acquisition were carried forward for rats that reached the criterion. The main findings were clear. The two Sham-lesion groups acquired their respective tasks, but not a single ATNlesion rat showed task acquisition irrespective of the inclusion of a 10-s trace between the odor and object stimuli. The primary evidence for acquisition is whether the rats learned to inhibit their response to the object in the non-rewarded trials (Figure 5a).
On this measure, the two Sham groups progressively delayed responding to the object on non-rewarded odor–object pairings during training (i.e. showed reduced reciprocal latency scores). By contrast, all rats in both ATN-lesion groups showed increasingly faster response latencies as training progressed. That is, on non-rewarded trials, the ATNlesion rats learned only to search more quickly under the object despite the odor–object pairing being incorrect. Note that the two ATN-lesion groups were, if anything, slower to respond than the Sham-lesion groups at the start of training on the non-rewarded trials, so general hyperactivity was not evident after ATN lesions in this task (Block 1 latency range: ATN-lesion 2.5–5.0 s, Shamlesion 1.9–4.5 s; Block 10: ATN-lesion 1.4–3.0 s, Shamlesion 5.5–7.6 s).
ANOVA of latencies on the nonrewarded trials produced a significant main effect across the four groups (F3,27 = 22.63, p < 0.001) and a significant Group Block interaction (F27,243 = 26.27, p < 0.001). The two Sham groups did not differ on this measure, and the two ATN groups did not differ, but both Sham groups differed from each of the ATN-lesion groups (p < 0.001).
When the odor–object pairing was rewarded, all four groups showed increasingly faster response latencies (i.e. showed increased reciprocal scores) across blocks of trials (Block main effect, F9,27 = 75.78, p < 0.001; Figure 5b), irrespective of Group (Group Block, F27,243 = 0.63, p = 0.92; Figure 5b). The Group main effect was again significant (F3,27 = 5.67, p = 0.003), which in this instance was due to faster running speed by rats in the Sham-Trace group compared with both ATN groups (p < 0.008) and the Sham-No-Trace rats (p = 0.02). However, the remaining three groups showed similar response latency on the rewarded trials. Again, note that the two ATN-lesion groups showed slower responses at the start of training on the rewarded trials, compared with the Sham-lesion groups.
The direct comparison across latencies for nonrewarded and rewarded trials is shown in Figure 5c, expressed as latency difference scores. Figure 5c clearly shows that no learning occurred in either ATN-lesion group, whereas both Sham-lesion groups acquired the task (Group Block interaction, F27,243 = 23.99, p < 0.001). The latency difference score suggests that the Sham-Trace group acquired the task more strongly than the Sham-No Trace group, and the mean differences between these two groups were significant for the last three blocks of trials (p < 0.01). This difference between the two sham groups, however, was primarily driven by their differences in trials with rewarded stimulus pairings (compare Figure 5b with Figure 5a).
3.5 | Retention test
All groups showed a response on the non-rewarded session at the 5-day retention test that was similar to their corresponding performance on Block 10 of training [Group Block (Retention Test vs. Block 10), F3,27 = 2.00, p = 0.13; Figure 5a]. Despite the nonsignificant Group Block interaction, there was an overall difference in latency across Block 10 versus the retention test (Retention vs. Block 10, F1,27 = 7.22, p = 0.01) that seems mostly due to ATN-lesion rats responding faster in the retention test. A significant group main effect (F3,27 = 73.13, p < 0.001) was driven by Sham groups continuing to inhibit responding on the non-rewarded trials, unlike the ATN-lesion groups (each Sham group vs. each ATN-lesion group, p < 0.001).
For rewarded trials, all four groups showed similar mean response latencies in the retention test compared with latency in Block 10 of training (Block, F1,27 = 3.75, p = 0.06; Group Block, F3,27 = 1.02, p = 0.39). Faster running speeds in the Sham-Trace group than all other groups were evident across Block 10 and the retention test (Group, F3,27 = 3.85, p = 0.02; Sham-Trace vs. all other groups, p < 0.03).
The direct comparison of rewarded and non-rewarded trials on the retention test, expressed as a latency difference score (Figure 5c), confirmed that all four groups maintained similar performance on this measure compared to their Block 10 acquisition performance (Group, F3,27 = 189.43, p < 0.001; Block, F1,27 = 0.08, p = 0.76; Group Block interaction, F3,27 = 0.22, p = 0.88).
3.6 | Zif268 expression
Our primary interest was to determine group differences in the regions of interest. Given the behavioral outcomes obtained, we do not report associations between Zif268 expression and performance as they would produce spurious correlations that are driven by the clear non-lesion versus sham status differences and small within-group variation on the primary measure of interest, that is, latencies on non-rewarded trials. Similarly, any associations evident in regions such as the retrosplenial cortex would again create an artificial correlation due to the marked group-level differences in Zif268 expression.
3.7 | Prefrontal regions
Zif268 expression across the four groups in the prelimbic prefrontal cortex (A32V) and the anterior cingulate cortex (A32D; A24b; A24a) is shown in Figure 6. There was a significant Group main effect for Area 32V (F2,37 = 5.49, p < 0.01; Figure 6b). Here, the two ATN groups showed similar overall expression to each other (p = 0.32), as well as the Sham-No Trace group (p > 0.1), but both lesion groups showed lower expression than in the Sham-Trace group (p < 0.02). The lower Zif268 expression in the Sham-No Trace group compared with the Sham-Trace group did not reach significance (p = 0.06). There was also a significant main effect across the four Layers (F3,81 = 147.8, p < 0.001), but no Group Layer interaction (F9,181 = 1.15, p > 0.3).
For the anterior cingulate (Cg) regions (Figure 6c), the two ATN-lesion groups showed lower Zif268 expression than both of the two Sham-lesion groups (Group F3,27 = 12.47, p < 0.001; both Sham groups differed from each of the ATN groups, p < 0.004, but not from each other, p = 0.31). A significant Group Layer interaction (F6,54 = 4.02, p < 0.002) reflected larger differences between ATN and Sham groups for Layer II and Layer III (p < 0.001) than for Layer V, where the groups did not differ significantly (p > 0.1). Expression differed across the three cingulate regions (Region main effect, F2,54 = 11.41, p < 001), being lowest for A24a (posterior) compared with both A32D and A24b (p < 0.002), and was highest for Layer III (Layer, F2,54 = 242.38, p < 0.001), but the size of these differences varied between layers across the three cingulate regions (Region Layer, F4,108 = 5.29, p < 0.001). However, the Group Region and Group Region Layer interactions were nonsignificant (all F < 1.0).
3.8 | Hippocampal and parahippocampal regions
Figure 7 shows the Zif268 expression for the four groups in the hippocampal and parahippocampal regions. For the hippocampus, the dorsal CA1 and CA3 subregions, especially CA1, showed higher expression than the ventral hippocampal CA1 and CA3 (dorsal vs. ventral, F1,24 = 541.4, p < 0.001; CA1 vs. CA3, F1,24 = 759.3, p < 0.001; interaction between these two factors, F1,24 = 428.4, p < 0.001). The most interesting finding, however, was that there was a higher expression in the dorsal CA1 in the Sham-Trace group than in each of the other three groups (p < 0.02), supported by a significant triple interaction for Group [dorsal region vs. ventral region] [CA1 vs. CA3], F3,24 = 6.09, p < 0.003). For the other three groups, dorsal CA1 expression was the lowest in the ATN-Trace group (ATN-Trace vs. Sham-No Trace and ATN-No Trace, p < 0.03), whereas the two No Trace groups did not differ significantly (p = 0.84). Analysis of the dorsal dentate gyrus (DG; only dorsal was examined) showed higher expression in the hilus than the granular cell layer (F1,27 = 386.1, p < 0.001), which was greater for the Sham-Trace and ATN-No Trace groups than for the Sham-No Trace and ATN-Trace conditions (Group DG Subregion, F3,27 = 4.07, p < 0.01). The ventral subiculum showed higher expression than the dorsal subiculum (F1,27 = 9.02, p < 0.005), but there was no Group effect (F3,27 = 1.41, p = 0.25) or Group [dorsal vs. ventral regions] interaction (F3,27 = 0.15, p = 0.92). In the parahippocampal regions, the perirhinal cortex showed lower expression than the two entorhinal cortex areas (F2,54 = 16.2, p < 0.001), but the groups did not differ at these sites (Group, F3,27 < 1.0; Group Region, F6,54 = 1.8, p > 0.1).
3.9 | Retrosplenial cortex
Figure 8 shows the Zif268 expression in the superficial and deep layers of Rga, Rgb, and Rdg. The two ATN groups showed markedly lower expression across these three regions than was shown by the two Sham groups (Group main effect, F3.27 = 40.9, p < 0.001). For the aggregated values across the three regions, the two Sham groups had higher mean Zif268 values than both ATNlesion groups (p = 0.0001), but the Sham-Trace group also showed higher levels than the Sham-No Trace group (p = 0.04). The difference between Sham groups and ATN groups was smallest for the Riga region (Group Region, F6,54 = 2.72, p < 0.02). The level of Zif268 expression between groups also varied across layers (Group Layer, F3,17 = 40.0, p < 0.001). This interaction reflected differences between both Sham groups compared with both ATN groups that were larger in the superficial layers than in the deep layers. Nonetheless, the Sham versus ATN group effects were still significant in the deep layers (p < 0.001). In addition, the Sham-Trace group showed higher expression in the superficial layers compared with the Sham-No Trace group (p = 0.03), whereas expression in the deep layers did not differ between these two groups (p = 0.23). There was no Group Region Layer interaction (F6,54 = 1.0, p = 0.43).
3.10 | Auditory control cortex
No differences were found in the auditory (control) cortex (Group, F3,27 = 1.2, p = 0.35).
4 | DISCUSSION
This study aimed to examine the effects of ATN lesions on non-spatial paired-associate memory and the influence of an explicit delay (i.e. a 10-s trace) between the presented odor and object stimuli. Evidence of impaired paired-associate memory after ATN lesions has previously been reported only when one of the paired components required the processing of distal spatial cues (Dumont et al., 2014; Gibb et al., 2006; Sziklas & Petrides, 1999).
We had anticipated that the effects of ATN lesions on non-spatial paired-associate memory in our task would be most evident when an explicit trace procedure was used. This was because it has been suggested that CA1 lesions only impair non-spatial paired-associate memory when a 10-s 'trace' is used (Kesner et al., 2005) and the microstructural integrity of CA1 neurons is reduced by both ATN lesions (Harland et al., 2014) and by mammillothalamic tract lesions that cause ATN dysfunction (Dillingham et al., 2019).
Moreover, the key examples of non-spatial memory impairment after ATN lesions examined temporal discriminations among multiple objects or odor items presented within a single block of trials (Dumont & Aggleton, 2013; Wolff et al., 2006). However, none of the 17 rats with ATN lesions showed evidence of acquisition of the odor–object paired-associate task, including those not trained with an explicit delay between the non-spatial stimuli. Despite an extended period of training, the ATNlesion rats were unable to show inhibited response to the non-rewarded odor–object pairings. General hyperactivity in the ATN-lesion rats does not seem a feature of this impairment, because they showed a slower response than the sham-lesion rats during the initial stages of acquisition.
The Sham-Trace group showed shorter latencies than the other three groups on the rewarded trials, but this difference may reflect increased anticipation of reward when restrained for a 10-s delay rather than a measure of faster acquisition by this group. The faster acquisition would be expected to be reflected by response inhibition on the non-rewarded trials, but the two sham lesion groups did not differ on this measure.
The failure to learn the paired-associate memory tasks after ATN lesions does not appear to be due to poor inhibition or impaired sensory processing. The ATNlesion rats demonstrated rapid acquisition in both the simple object discrimination task and the simple odor discrimination task, which was equal to that shown by sham-lesion rats. The task demands for these simple discriminations were identical to the paired-associate task and used the same apparatus. For the same reason, the paired-associate deficit after ATN lesions is also unlikely to be due to a simple failure of attention to the individual stimuli used. ATN lesions impair the ability to learn an attentional set and facilitate extradimensional shifts but do not change sustained attention or behavioral flexibility (Chudasama & Muir, 2001; Kinnavane et al., 2019; Wright et al., 2015). The rapid acquisition of the simple discriminations in the current runway task contrasted with slower acquisition when we trained a previous group of rats on an open circular platform to learn simple odor discrimination and, especially, simple object discrimination (Bell, 2007). So, it is possible that the use of a runway and the explicit reduction of distracting spatial cues, plus the active interaction with the object to search for food, facilitated attention to the non-spatial stimuli in the current study.
The severity of the lesion impairment for learning the association between odor and object stimuli suggests that this task is strongly dependent on the integrity of the ATN. This evidence is at odds with the suggestion that paired-associate impairments after ATN lesions require the use of multi-modal spatial stimuli (Dumont et al., 2014; Nelson, 2021). One possible explanation for the difference between the outcome in the current study and that of Dumont et al. (2014) is that discrete nonspatial stimuli in biconditional discrimination learning tasks, such as specific objects or odors, may pose a greater attentional demand on the establishment of a unique integrated representation compared with the use of a general local context, such as the thermal, visual or tactile local environment. In a similar way, the profound susceptibility of spatial paired-associate tasks to impairment after ATN lesions may also rely on the integration of relational spatial cues in combination with a salient discrete cue because the acquisition of a simple spatial discrimination per se was only partially impaired (Dumont et al., 2014; Gibb et al., 2006). One unexpected finding is that rats with ATN lesions showed no deficit when they were required to select a particular location in a cross maze based on a conditional visual cue at a choice point (Sziklas & Petrides, 2007). In that situation, however, the conditional relationship was determined by a single salient cue presented in a single place that had no ambiguity or embedded association with the different positions of the correct spatial location. This contrasts with a complete failure by ATN-lesion rats when they had to learn an object-place association in which they had to select one of two correct objects based on their associated location (Sziklas & Petrides, 1999).
The finding that ATN lesions produced a profound deficit in odor–object paired-associate memory, irrespective of the presence of a 10-s trace between the stimuli, adds to our earlier evidence of a deficit of paired-associate memory when the object and odor were presented simultaneously on a cheeseboard platform (Bell, 2007). Together, this suggests that the memory deficits after ATN lesions on odor–object paired-associate memory is an additional example that ATN lesions do not always mirror the pattern of conditional associative memory deficits produced by lesions to the hippocampal system (Sziklas & Petrides, 2004, 2007).
Whereas ATN lesions can cause greater spatial memory impairments than fornix lesions (Warburton & Aggleton, 1999), or deficits in object–place and geometric discrimination tasks that are not found with fornix lesions (Aggleton et al., 2009; Sziklas et al., 1998), there is less evidence that ATN lesions can produce severe memory impairments in tasks that are generally unaffected by lesions of the hippocampal formation. Concerning non-spatial arbitrary associations, the evidence that the hippocampus is only critical when a 10-s trace is used between the two stimuli was derived by comparing acquisition in two different tasks. Gilbert and Kesner (2002) reported that large hippocampal lesions did not impair object–odor–paired associate memory when tested on a cheeseboard platform in which the two stimuli were presented simultaneously. In a similar runway to ours, but with an object presented before exposure to odourised sand that might contain a reward, Kesner et al. (2005) showed that dorsal CA1 lesions but not CA3 lesions produced a deficit when using a 10-s trace condition.
Neither we nor Kesner and colleagues examined the effects of hippocampal lesions in the runway task without the trace condition. So, we cannot be sure that rats with hippocampal lesions would be unimpaired under the no-trace condition when trained on the runway using our procedures. Nonetheless, our study extends the predicted association between CA1 function and temporal processing in a non-spatial paired-associate task by finding increased Zif268 expression in dorsal CA1 in the Sham-lesion group trained using a 10-s trace relative to the Sham-lesion No Trace group. By contrast, the mean Zif268 expression in the dorsal CA1 was lowest in the ATN-lesion Trace group.
There was also evidence, albeit weaker, that the trace condition in sham lesion rats was associated with increased Zif268 expression in the superficial layers of the retrosplenial cortex. The pattern of different behavioral performances in the lesion and non-lesion groups at retention made it inappropriate to examine the association between variations in performance and variations in Zif268 expression. As in previous studies (Aggleton & Nelson, 2015; Perry et al., 2018), the strongest effect was a marked reduction in IEG expression after ATN lesions in the retrosplenial cortex, especially the superficial layers. It seems likely that this finding is due to the loss of or diminished activity in the direct inputs from the ATN to the RSC (Barnett et al., 2021).
There has been growing awareness that ATN lesions may exert influences beyond spatial memory impairments (Nelson, 2021; Wolff et al., 2006). One example is when ATN lesions slow the acquisition of non-spatial attentional-set learning, which may be due to a functional relationship between the ATN and the midcingulate regions of the cingulate cortex rather than medial prefrontal cortex connections (Bubb et al., 2021; Wright et al., 2015). To our knowledge, however, this attentional-set task has not been examined with hippocampal lesions in rats. A clearer example of a dissociation between ATN lesions and lesions of the hippocampus is that only the former injury impairs attentional processes associated with latent inhibition (Nelson et al., 2018). It is possible that an impoverished ability to establish the relevance or predictiveness of stimulus–stimulus associations provides a unifying account not only of attentional set learning and latent inhibition (see Nelson et al., 2018) but also of instances of impaired paired-associate learning after ATN lesions. Rather than ascribing the role of the ATN from the perspective of either hippocampal (for space and time) or frontal (for attentional) processes, the broader implication is that the ATN may support memory processing by actively orchestrating attention to certain classes of stimulus–stimulus associations and their representation across multiple brain structures (Leszcynski & Staudigl, 2016). Precisely how to characterize the classes of impaired memory remains an experimental challenge for the future. What is clear, from the current study, is that explanations based only on a spatial/non-spatial dichotomy are unable to account for profound memory deficits that can be found in both domains after ATN lesions.
Our findings bring the effects of ATN lesions in rats closer into line with paired-associate memory impairments in clinical amnesia after injury to the mammillary body-ATN axis (Rempel-Clower et al., 1996; Squire et al., 2020). There is, however, a clear difference between the slow acquisition of paired-associate memory in intact rats and the rapid acquisition of paired-associate memory in humans with intact memory systems. Paired-associate tasks are assumed to reflect episodic-like memory by measuring the ability to form unique representations of multiple stimuli rather than memory for individual components (Crystal & Smith, 2014; Eichenbaum & Fortin, 2009). In our study, however, the intact rats required 4–5 weeks and over 300 training trials before there was clear evidence of acquisition, which suggests that the task could be more rule-based or semantic-like in these rats.
This limitation could be circumvented in future work by first training intact rats on one or more non-spatial paired-associate tasks before testing the acquisition of a new odor–object pairing after ATN lesions. In this way, the general rule of making an association would already be established, and, perhaps, the rate of acquisition would then be relatively rapid for a new task in intact rats. In addition, temporary chemogenetic or optogenetic manipulations of the ATN could provide an opportunity to investigate the impact that these nuclei have on retarding acquisition rather than preventing acquisition or on their impact on retention rather than acquisition. It would also be informative to learn whether many or only some neural projections from the ATN support this example of non-spatial paired-associate learning.
Neuroanatomical evidence suggests a different pattern of neural connections with limbic and cortical memory structures among the three ATN nuclei, that is, the anterodorsal nuclei, anteroventral nuclei, and the anteromedial nuclei (Bubb et al., 2017; Lomi et al., 2021; Nelson, 2021). In addition, these ATN component nuclei have different molecular and electrophysiological characteristics that may underpin different behavioral functions (Jankowski et al., 2013; Roy et al., 2021, 2022; Safari et al., 2020). Lesions or genetic manipulations of the individual nuclei of the ATN could provide insight as to whether paired-associate memory relies on one or more of the AD, AV, or AM. It may be the case that the impact of ATN lesions on this task preferentially involves frontal brain regions and/or the retrosplenial cortex and their respective involvement in rule-based and knowledge-based systems rather than event-based memory (Hunsaker & Kesner, 2018). These issues could also be addressed using contralateral disconnection lesions involving the ATN, as this experimental approach has been used to successfully demonstrate the system-wide influence of the ATN-hippocampal axis in spatial tasks (Dumont et al., 2010; Warburton et al., 2000, 2001).
The present study provides unequivocal evidence that ATN lesions produce substantial impairments in nonspatial paired-associate learning and memory, irrespective of the presence of an explicit temporal component. Extensive training demonstrated no evidence of learning in these ATN-lesion rats. The fact that there was relatively little to minimal damage to the immediately adjacent intralaminar or mediodorsal thalamic regions suggests that these impairments were specific to the ATN lesion. Evidence from these non-spatial paired-associate tasks suggests a new perspective on the role of the ATN as a critical node within the 'hippocampal–diencephalic– cingulate' memory network (Bubb et al., 2017). This strengthens the view that the ATN does not operate primarily as a relay for hippocampal information (Wolff & Vann, 2019). Instead, the ATN may actively control the generation of some classes of arbitrary memory representations in the brain.
ACKNOWLEDGEMENTS
This research was supported by University of Canterbury equipment and research grants and Early Career support (JJH) from Brain Research New Zealand – Rangahau Roro Aotearoa. Open access publishing facilitated by the University of Canterbury, as part of the Wiley - University of Canterbury agreement via the Council of Australian University Librarians.
REFERENCES
Aggleton, J. P., Amin, E., Jenkins, T. A., Pearce, J. M., & Robinson, J. (2011). Lesions in the anterior thalamic nuclei of rats do not disrupt the acquisition of stimulus sequence learning. The Quarterly Journal of Experimental Psychology, 64(1), 65– 73. https://doi.org/10.1080/17470218.2010.495407
Aggleton, J. P., & Brown, M. W. (1999). Episodic memory, amnesia, and the hippocampal–anterior thalamic axis. Behavioral and Brain Sciences, 22(3), 425–444. https://doi.org/10.1017/ s0140525x99002034
Aggleton, J. P., & Nelson, A. J. D. (2015). Why do lesions in the rodent anterior thalamic nuclei cause such severe spatial deficits? Neuroscience & Biobehavioral Reviews, 54, 131–144.https://doi.org/10.1016/j.neubiorev.2014.08.013
Aggleton, J. P., Poirier, G. L., Aggleton, H. S., Vann, S. D., & Pearce, J. M. (2009). Lesions of the fornix and anterior thalamic nuclei dissociate different aspects of hippocampal-dependent spatial learning: Implications for the neural basis of scene learning. Behavioral Neuroscience, 123(3), 504–519.https://doi.org/10.1037/a0015404
Barnett, S. C., Parr-Brownlie, L. C., Perry, B. A., Young, C. K., Wicky, H. E., Hughes, S. M., McNaughton, N., & Dalrymple-Alford, J. C. (2021). Anterior thalamic nuclei neurons sustain memory. Current Research in Neurobiology, 2, 100022. https:// doi.org/10.1016/j.crneur.2021.100022
Bell, R. (2007). Anterior and lateral thalamic lesions in object-odor paired associate learning (unpublished master's thesis). University of Canterbury, Christchurch, New Zealand.
Bubb, E. J., Aggleton, J. P., O'Mara, S. M., & Nelson, A. J. D. (2021). Chemogenetics Reveal an Anterior Cingulate–Thalamic Pathway for Attending to Task-Relevant Information. Cerebral Cortex, 31(4), 2169–2186. https://doi.org/10.1093/cercor/bhaa353
Bubb, E. J., Kinnavane, L., & Aggleton, J. P. (2017). Hippocampal– diencephalic–cingulate networks for memory and emotion: An anatomical guide. Brain and Neuroscience Advances, 1(1), 1–20. https://doi.org/10.1177/2398212817723443
For more information:1950477648nn@gmail.com
