Limits...
Enhancement of encoding and retrieval functions through theta phase-specific manipulation of hippocampus.

Siegle JH, Wilson MA - Elife (2014)

Bottom Line: Assessing the behavioral relevance of the hippocampal theta rhythm has proven difficult, due to a shortage of experiments that selectively manipulate phase-specific information processing.Using closed-loop stimulation, we triggered inhibition of dorsal CA1 at specific phases of the endogenous theta rhythm in freely behaving mice.Conversely, stimulation in the retrieval segment enhanced performance when inhibition was triggered by the trough of theta.

View Article: PubMed Central - PubMed

Affiliation: Picower Institute for Learning and Memory, Massachusetts Institute of Technology, Cambridge, United States Department of Brain and Cognitive Sciences, Massachusetts Institute of Technology, Cambridge, United States jsiegle@mit.edu.

No MeSH data available.


Related in: MedlinePlus

Behavioral modulation depends on both theta phase and tasksegment.(A) Illustration of the two manipulations performed in thisexperiment. On any given ‘non-baseline’ trial, stimulation wastriggered by the peak (purple phase) or trough (teal phase) of the4–12 Hz theta rhythm. The resulting light pulses recruited inhibitionfor ∼25 ms, or approximately 1/5 of the 125 ms theta cycle.(B) Accuracy relative to baseline for four mice in fourconditions: optogenetic stimulation triggered at the peak (purple) or trough(teal) of theta, in either the retrieval (left) or encoding (right) segmentsof the track. Mean ± SEM, with results for each mouse overlaid.Individual results in the gray regions are significantly different frombaseline (p<0.05, p.d.f. of binomial distribution with probabilityequal to baseline accuracy). (C) Same data as in b, butrepresented on the same axes. Note that peak-triggered stimulation in theencoding segment consistently improves performance more than the same typeof stimulation in the retrieval segment (points above diagonal line). Theopposite effects are seen for trough-triggered stimulation. (D)Schematic of all possible ‘double-dissociation’ scenarios usedfor establishing bootstrap significance levels of the actual result.(E) Performance on baseline (no stimulation) trials for fourdifferent trial types: (1) mice are cued to switch arms after a correctchoice (correct/switch), (2) mice are cued to return to the same arm after acorrect choice (correct/stay), (3) mice are cued to switch arms after anincorrect choice (incorrect/switch), and (4) mice are cued to return to thesame arm after an incorrect choice. Trials are grouped by retrievalstimulation or encoding stimulation conditions. For both conditions,changing trial type has a significant effect on performance: retrievalstimulation, χ2 = 8.4, p=0.038;encoding stimulation, χ2 = 8.1,p=0.044; Friedman test (nonparametric, repeated-measures ANOVA).(F) Change in performance with the addition of closed-loopoptogenetic stimulation for the four trial types in E.DOI:http://dx.doi.org/10.7554/eLife.03061.007
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4384761&req=5

fig5: Behavioral modulation depends on both theta phase and tasksegment.(A) Illustration of the two manipulations performed in thisexperiment. On any given ‘non-baseline’ trial, stimulation wastriggered by the peak (purple phase) or trough (teal phase) of the4–12 Hz theta rhythm. The resulting light pulses recruited inhibitionfor ∼25 ms, or approximately 1/5 of the 125 ms theta cycle.(B) Accuracy relative to baseline for four mice in fourconditions: optogenetic stimulation triggered at the peak (purple) or trough(teal) of theta, in either the retrieval (left) or encoding (right) segmentsof the track. Mean ± SEM, with results for each mouse overlaid.Individual results in the gray regions are significantly different frombaseline (p<0.05, p.d.f. of binomial distribution with probabilityequal to baseline accuracy). (C) Same data as in b, butrepresented on the same axes. Note that peak-triggered stimulation in theencoding segment consistently improves performance more than the same typeof stimulation in the retrieval segment (points above diagonal line). Theopposite effects are seen for trough-triggered stimulation. (D)Schematic of all possible ‘double-dissociation’ scenarios usedfor establishing bootstrap significance levels of the actual result.(E) Performance on baseline (no stimulation) trials for fourdifferent trial types: (1) mice are cued to switch arms after a correctchoice (correct/switch), (2) mice are cued to return to the same arm after acorrect choice (correct/stay), (3) mice are cued to switch arms after anincorrect choice (incorrect/switch), and (4) mice are cued to return to thesame arm after an incorrect choice. Trials are grouped by retrievalstimulation or encoding stimulation conditions. For both conditions,changing trial type has a significant effect on performance: retrievalstimulation, χ2 = 8.4, p=0.038;encoding stimulation, χ2 = 8.1,p=0.044; Friedman test (nonparametric, repeated-measures ANOVA).(F) Change in performance with the addition of closed-loopoptogenetic stimulation for the four trial types in E.DOI:http://dx.doi.org/10.7554/eLife.03061.007

Mentions: The effects of closed-loop optogenetic feedback on behavior depended on both thephase of theta used to trigger stimulation and the region of the track in which thestimulation occurred. On individual trials, 10 ms light pulses were triggered oneither the peak or trough of theta (Figure 5A,phase relative to theta at the hippocampal fissure). When stimulation occurred in theretrieval segment, performance did not differ between baseline and peak-triggeredstimulation for 4/4 mice (mean of 57.3 ± 10.0% correct for baseline vs 57.8± 10.5% correct for peak, individual results in Table 1). For trough-triggered stimulation, however, performanceimproved significantly in 4/4 mice (71.0 ± 8.2% correct for trough; significancedetermined by the p.d.f. of the binomial distribution, with baseline accuracy foreach mouse used as the ‘chance’ level). The opposite effects wereobserved for stimulation in the encoding segment. In this condition, performanceduring trials with trough-triggered stimulation did not differ from baseline in 3/4mice (mean of 59.1 ± 2.4% correct for baseline vs 58.7 ± 10.5% correct fortrough; 1 mouse had significantly impaired performance in the trough-stimulationcondition). For peak-triggered stimulation, performance improved significantly in 3/4mice (mean of 69.6 ± 14.7% correct; 1 mouse showed no difference frombaseline).10.7554/eLife.03061.007Figure 5.Behavioral modulation depends on both theta phase and tasksegment.


Enhancement of encoding and retrieval functions through theta phase-specific manipulation of hippocampus.

Siegle JH, Wilson MA - Elife (2014)

Behavioral modulation depends on both theta phase and tasksegment.(A) Illustration of the two manipulations performed in thisexperiment. On any given ‘non-baseline’ trial, stimulation wastriggered by the peak (purple phase) or trough (teal phase) of the4–12 Hz theta rhythm. The resulting light pulses recruited inhibitionfor ∼25 ms, or approximately 1/5 of the 125 ms theta cycle.(B) Accuracy relative to baseline for four mice in fourconditions: optogenetic stimulation triggered at the peak (purple) or trough(teal) of theta, in either the retrieval (left) or encoding (right) segmentsof the track. Mean ± SEM, with results for each mouse overlaid.Individual results in the gray regions are significantly different frombaseline (p<0.05, p.d.f. of binomial distribution with probabilityequal to baseline accuracy). (C) Same data as in b, butrepresented on the same axes. Note that peak-triggered stimulation in theencoding segment consistently improves performance more than the same typeof stimulation in the retrieval segment (points above diagonal line). Theopposite effects are seen for trough-triggered stimulation. (D)Schematic of all possible ‘double-dissociation’ scenarios usedfor establishing bootstrap significance levels of the actual result.(E) Performance on baseline (no stimulation) trials for fourdifferent trial types: (1) mice are cued to switch arms after a correctchoice (correct/switch), (2) mice are cued to return to the same arm after acorrect choice (correct/stay), (3) mice are cued to switch arms after anincorrect choice (incorrect/switch), and (4) mice are cued to return to thesame arm after an incorrect choice. Trials are grouped by retrievalstimulation or encoding stimulation conditions. For both conditions,changing trial type has a significant effect on performance: retrievalstimulation, χ2 = 8.4, p=0.038;encoding stimulation, χ2 = 8.1,p=0.044; Friedman test (nonparametric, repeated-measures ANOVA).(F) Change in performance with the addition of closed-loopoptogenetic stimulation for the four trial types in E.DOI:http://dx.doi.org/10.7554/eLife.03061.007
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4384761&req=5

fig5: Behavioral modulation depends on both theta phase and tasksegment.(A) Illustration of the two manipulations performed in thisexperiment. On any given ‘non-baseline’ trial, stimulation wastriggered by the peak (purple phase) or trough (teal phase) of the4–12 Hz theta rhythm. The resulting light pulses recruited inhibitionfor ∼25 ms, or approximately 1/5 of the 125 ms theta cycle.(B) Accuracy relative to baseline for four mice in fourconditions: optogenetic stimulation triggered at the peak (purple) or trough(teal) of theta, in either the retrieval (left) or encoding (right) segmentsof the track. Mean ± SEM, with results for each mouse overlaid.Individual results in the gray regions are significantly different frombaseline (p<0.05, p.d.f. of binomial distribution with probabilityequal to baseline accuracy). (C) Same data as in b, butrepresented on the same axes. Note that peak-triggered stimulation in theencoding segment consistently improves performance more than the same typeof stimulation in the retrieval segment (points above diagonal line). Theopposite effects are seen for trough-triggered stimulation. (D)Schematic of all possible ‘double-dissociation’ scenarios usedfor establishing bootstrap significance levels of the actual result.(E) Performance on baseline (no stimulation) trials for fourdifferent trial types: (1) mice are cued to switch arms after a correctchoice (correct/switch), (2) mice are cued to return to the same arm after acorrect choice (correct/stay), (3) mice are cued to switch arms after anincorrect choice (incorrect/switch), and (4) mice are cued to return to thesame arm after an incorrect choice. Trials are grouped by retrievalstimulation or encoding stimulation conditions. For both conditions,changing trial type has a significant effect on performance: retrievalstimulation, χ2 = 8.4, p=0.038;encoding stimulation, χ2 = 8.1,p=0.044; Friedman test (nonparametric, repeated-measures ANOVA).(F) Change in performance with the addition of closed-loopoptogenetic stimulation for the four trial types in E.DOI:http://dx.doi.org/10.7554/eLife.03061.007
Mentions: The effects of closed-loop optogenetic feedback on behavior depended on both thephase of theta used to trigger stimulation and the region of the track in which thestimulation occurred. On individual trials, 10 ms light pulses were triggered oneither the peak or trough of theta (Figure 5A,phase relative to theta at the hippocampal fissure). When stimulation occurred in theretrieval segment, performance did not differ between baseline and peak-triggeredstimulation for 4/4 mice (mean of 57.3 ± 10.0% correct for baseline vs 57.8± 10.5% correct for peak, individual results in Table 1). For trough-triggered stimulation, however, performanceimproved significantly in 4/4 mice (71.0 ± 8.2% correct for trough; significancedetermined by the p.d.f. of the binomial distribution, with baseline accuracy foreach mouse used as the ‘chance’ level). The opposite effects wereobserved for stimulation in the encoding segment. In this condition, performanceduring trials with trough-triggered stimulation did not differ from baseline in 3/4mice (mean of 59.1 ± 2.4% correct for baseline vs 58.7 ± 10.5% correct fortrough; 1 mouse had significantly impaired performance in the trough-stimulationcondition). For peak-triggered stimulation, performance improved significantly in 3/4mice (mean of 69.6 ± 14.7% correct; 1 mouse showed no difference frombaseline).10.7554/eLife.03061.007Figure 5.Behavioral modulation depends on both theta phase and tasksegment.

Bottom Line: Assessing the behavioral relevance of the hippocampal theta rhythm has proven difficult, due to a shortage of experiments that selectively manipulate phase-specific information processing.Using closed-loop stimulation, we triggered inhibition of dorsal CA1 at specific phases of the endogenous theta rhythm in freely behaving mice.Conversely, stimulation in the retrieval segment enhanced performance when inhibition was triggered by the trough of theta.

View Article: PubMed Central - PubMed

Affiliation: Picower Institute for Learning and Memory, Massachusetts Institute of Technology, Cambridge, United States Department of Brain and Cognitive Sciences, Massachusetts Institute of Technology, Cambridge, United States jsiegle@mit.edu.

No MeSH data available.


Related in: MedlinePlus