Journal Club - "Basal ganglia and cerebellar lesions causally impact the neural encoding of temporal regularities"

This slide presentation provides an overview of the highlights of the journal article by Criscuolo et al. 2025 published in Imaging Neuroscience.

1. Journal Club – Music and Neuroscience Lab
Ana Luísa Pinho
27
th
of October, 2025

2. Background (1/2)

3. Background (1/2)
Timing is the neural encoding of when events occur and
the use of that information to guide perception and action.

4. Background (1/2)
Timing is the neural encoding of when events occur and
the use of that information to guide perception and action.
Temporal regularities

Many sounds occur at regular intervals (e.g., rhythmic sequences).

Such regularity supports timing by enabling predictions of upcoming onsets.

5. Background (1/2)
Timing is the neural encoding of when events occur and
the use of that information to guide perception and action.
Temporal regularities

Many sounds occur at regular intervals (e.g., rhythmic sequences).

Such regularity supports timing by enabling predictions of upcoming onsets.
Encoding & synchronization via predictions

Temporal predictions drive anticipatory activity and phase-aligned oscillations,
letting the brain synchronize to regular input.

6. Background (2/2)

7. Background (2/2)
Sub-cortical neurocognitive roles

Cerebellum (CE): precise onset/interval timing stable timing representations.

Basal Ganglia (BG): generate temporal predictions and use relative timing to extract the beat

progressive alignment of expectations across sequences

Both operate within an extended motor network recruited for predictive, synchronized behavior.

8. Background (2/2)
Sub-cortical neurocognitive roles

Cerebellum (CE): precise onset/interval timing stable timing representations.

Basal Ganglia (BG): generate temporal predictions and use relative timing to extract the beat

progressive alignment of expectations across sequences

Both operate within an extended motor network recruited for predictive, synchronized behavior.
Gap

Causal comparison of CE vs BG within the same paradigm are scarce.

Most evidence comes from separate tasks/cohorts or correlational designs.

9. 
Provide a causal, side-by-side test of CE vs BG by examining how focal lesions affect neural
encoding and synchronization to temporal regularities during passive listening.

Quantify entrainment with Event-
­Related Potentials
(ERPs),
delta-band Inter-
­Trial Phase Coherence
(ITPC) and time-resolved ITPC (t-ITPC).

Introduce an acceleration-based delta stability metric indexing how steadily neural activity tunes to
and remains at the stimuli frequency.
Goals

10. Participants & Lesions
Groups & Demographics

N=33: CE, n=11; BG, n=11;

Healthy controls (HC), n=11.

Right‑handed,
non‑musicians;

matched on age/education;

no hearing/psychiatric
disorders.
Fig1. Brain lesion delineation and overlap across participants on the anatomical template.

11. Task & Stimuli
Fig. 2A: Experimental design
(sequence timing,
Standard and Deviant (STD/DEV) tones).
Design & Sounds

Passive listening to
96 isochronous sequences
at Stimulus Frequency (Sf) = 1.54 Hz
(Inter-Stimulus Onset (ISO) = 650 ms).

Each sequence: 13–16 tones
(pitch 400 Hz, duration 50 ms, loudness 70 dB),
with 1–2 amplitude-deviants (66 dB).

Trial: Fixation = 500 ms Sequence (8.45 – 10.4 s) Inter-Trial Interval = 2000 ms;
→ →
2 blocks ~10 min/each; report # softer tones at end.

12. Overview of the EEG Metrics (1/2)

13. Overview of the EEG Metrics (1/2)

ERPs (N100): amplitude, latency, trial-level
variance of both (tones 3–7).
1

14. Overview of the EEG Metrics (1/2)

ERPs (N100): amplitude, latency, trial-level
variance of both (tones 3–7).

More variance, especially in latency,
means less precise timing.
1

15. Overview of the EEG Metrics (1/2)

ERPs (N100): amplitude, latency, trial-level
variance of both (tones 3–7).

More variance, especially in latency,
means less precise timing.

ITPC @ 1.5 Hz: phase-locking at the stimulus
rate (delta band).
1
2

16. Overview of the EEG Metrics (1/2)

ERPs (N100): amplitude, latency, trial-level
variance of both (tones 3–7).

More variance, especially in latency,
means less precise timing.

ITPC @ 1.5 Hz: phase-locking at the stimulus
rate (delta band).

Higher ITPC: stronger alignment of delta-phase
w/ tone sequence.

Lower ITPC: weaker alignment.
1
2

17. Overview of the EEG Metrics (1/2)

ERPs (N100): amplitude, latency, trial-level
variance of both (tones 3–7).

More variance, especially in latency,
means less precise timing.

ITPC @ 1.5 Hz: phase-locking at the stimulus
rate (delta band).

Higher ITPC: stronger alignment of delta-phase
w/ tone sequence.

Lower ITPC: weaker alignment.

t-ITPC slope: build-up of phase alignment
across the sequence.
1
2

18. Overview of the EEG Metrics (1/2)

ERPs (N100): amplitude, latency, trial-level
variance of both (tones 3–7).

More variance, especially in latency,
means less precise timing.

ITPC @ 1.5 Hz: phase-locking at the stimulus
rate (delta band).

Higher ITPC: stronger alignment of delta-phase
w/ tone sequence.

Lower ITPC: weaker alignment.

t-ITPC slope: build-up of phase alignment
across the sequence.

Positive slope: phase alignment stronger as
the rhythm unfolds.

Flatter slope: no build-up of alignment.
1
2

19. Overview of the EEG Metrics (2/2)

20. Overview of the EEG Metrics (2/2)

Delta stability:
acceleration-based tuning steadiness of
instantaneous frequency near 1.5 Hz.
3
Fig 4. A

21. Overview of the EEG Metrics (2/2)

Delta stability:
acceleration-based tuning steadiness of
instantaneous frequency near 1.5 Hz.

High-stability:
delta signal steadily aligns to the beat.

Low stability: wobbly inst. frequency trace.
3
Fig 4. A

22. Fig 2.
ERP Results (1/3)

All groups show canonical auditory ERPs
(clear N100/P200 morphology)
during the sequence.

Cerebellar (CE) group shows visibly
reduced N100 amplitude vs controls;
BG is intermediate.

23. ERP Results (2/3)
Fig 2.

Averaging ERPs at tones 3–7 isolates
the steady portion (after onset transients).

Group difference in N100 size persists here:
CE < HC;
average latency similar across groups.

24. ERP Results (3/3)
Fig 2.

Amplitude:
CE < HC (significant);
BG not reliably different from HC.

Latency:
group means similar (no reliable group effect).

Latency variance (tone level):

↑
in CE & BG vs HC (less precise timing);


Amplitude variance (tone level):

↓
in lesion groups vs HC (weaker response)

25. Fig 3.
ITPC @ 1.5 Hz Results

All groups show a peak at 1.5 Hz, but HC > CE & BG (ANOVA p=.01; HC>CE p=.04; HC>BG p=.02).

No group differences in an unrelated band (alpha).

26. Fig 3.
ITPC @ 1.5 Hz Results

All groups show a peak at 1.5 Hz, but HC > CE & BG (ANOVA p=.01; HC>CE p=.04; HC>BG p=.02).

No group differences in an unrelated band (alpha).
Take-home:
Patients have
reduced delta phase-locking
at the beat.

27. Fig 3.
t-ITPC Results

HC show a rising time course (positive slope) from tone 3 onward;

Both patient groups show a shallower/flat slope (Kruskal-Wallis significant; HC > CE & BG in slope).

28. Fig 3.
t-ITPC Results

HC show a rising time course (positive slope) from tone 3 onward;

Both patient groups show a shallower/flat slope (Kruskal-Wallis significant; HC > CE & BG in slope).
Take-home:
Patients have
weaker build-up of
phase alignment
across the sequence.

29. Delta-dynamics Stability Results

Stability (S):
Cerebellar (CE) < Controls (HC) (significant);
BG HC (ns).
≈

Deviation (D): no reliable group differences.

Checks: no group differences in mean IF,
mean acceleration,
or latency to tuning.
Fig 4.

30. Delta-dynamics Stability Results

Stability (S):
Cerebellar (CE) < Controls (HC) (significant);
BG HC (ns).
≈

Deviation (D): no reliable group differences.

Checks: no group differences in mean IF,
mean acceleration,
or latency to tuning.
Take-home:
Beyond reduced phase-locking,
cerebellar damage specifically
disrupts how steadily the delta
rhythm stays tuned to the beat.
Fig 4.

31. Main Conclusions

32. 
Early auditory encoding is less precise in both lesion groups (higher N100 latency variance),
and N100 amplitude is specifically reduced in CE.
Main Conclusions

33. 
Early auditory encoding is less precise in both lesion groups (higher N100 latency variance),
and N100 amplitude is specifically reduced in CE.

Both cerebellar (CE) and basal-ganglia (BG) lesions reduce delta-band phase-locking
(lower ITPC, flatter t-ITPC build-up).
Main Conclusions

34. 
Early auditory encoding is less precise in both lesion groups (higher N100 latency variance),
and N100 amplitude is specifically reduced in CE.

Both cerebellar (CE) and basal-ganglia (BG) lesions reduce delta-band phase-locking
(lower ITPC, flatter t-ITPC build-up).

The novel delta stability metric isolates a CE-specific deficit,
showing that the delta rhythm locks less steadily near the beat (~1.5 Hz) in cerebellar lesions.
Main Conclusions

35. 
Early auditory encoding is less precise in both lesion groups (higher N100 latency variance),
and N100 amplitude is specifically reduced in CE.

Both cerebellar (CE) and basal-ganglia (BG) lesions reduce delta-band phase-locking
(lower ITPC, flatter t-ITPC build-up).

The novel delta stability metric isolates a CE-specific deficit,
showing that the delta rhythm locks less steadily near the beat (~1.5 Hz) in cerebellar lesions.

Together, the results causally implicate both CE and BG in neural entrainment
to temporal regularities, with cerebellum uniquely linked to steady tuning.
Main Conclusions

36. Discussion

37. Discussion

The study links subcortical damage to specific entrainment deficits,
moving beyond correlational to a causal lesion evidence.
Highlights

38. Discussion

The study links subcortical damage to specific entrainment deficits,
moving beyond correlational to a causal lesion evidence.

The stability metric adds a dynamics-level view that complements ITPC/t-ITPC and
reveals a cerebellar signature.
Highlights

39. Discussion

The study links subcortical damage to specific entrainment deficits,
moving beyond correlational to a causal lesion evidence.

The stability metric adds a dynamics-level view that complements ITPC/t-ITPC and
reveals a cerebellar signature.

Passive, isochronous listening shows that explicit timing demands are not required
to observe CE/BG contributions.
Highlights

40. Discussion

The study links subcortical damage to specific entrainment deficits,
moving beyond correlational to a causal lesion evidence.

The stability metric adds a dynamics-level view that complements ITPC/t-ITPC and
reveals a cerebellar signature.

Passive, isochronous listening shows that explicit timing demands are not required
to observe CE/BG contributions.
Highlights
Limitations

Small samples and heterogeneous lesions limit regional specificity and power.

41. Discussion

The study links subcortical damage to specific entrainment deficits,
moving beyond correlational to a causal lesion evidence.

The stability metric adds a dynamics-level view that complements ITPC/t-ITPC and
reveals a cerebellar signature.

Passive, isochronous listening shows that explicit timing demands are not required
to observe CE/BG contributions.
Highlights
Limitations

Small samples and heterogeneous lesions limit regional specificity and power.

Passive, strictly isochronous sequences may underplay BG-specific effects that emerge with
explicit beat inference or complex rhythms.

42. Discussion

The study links subcortical damage to specific entrainment deficits,
moving beyond correlational to a causal lesion evidence.

The stability metric adds a dynamics-level view that complements ITPC/t-ITPC and
reveals a cerebellar signature.

Passive, isochronous listening shows that explicit timing demands are not required
to observe CE/BG contributions.
Highlights
Limitations

Small samples and heterogeneous lesions limit regional specificity and power.

Passive, strictly isochronous sequences may underplay BG-specific effects that emerge with
explicit beat inference or complex rhythms.

Analyses emphasize delta; beta-band and source-level interactions (cortico-subcortical loops)
remain to be tested.

43. Appendix