Figure 1.
An example of oscillations in the size of the pupils of a urethane-anaesthetised rat.
Top: frames of the left and right eyes taken at different recording times. Dotted ellipses indicate detected edges of the pupils, and the corresponding area of the pupil is given in the bottom right corner of each frame. Bottom: graphs showing the oscillatory changes of the areas of pupils observed simultaneously in the left (red line) and right (blue line) eyes. Both graphs have the same time scale. Dotted lines with arrows indicate the exact times at which the frames shown in the top were taken.
Figure 2.
An example showing the two infra-slow components of oscillatory changes of the pupil size and its coincidence with cyclic alternations of brain state.
Top: two diagrams showing the rhythmic changes simultaneously observed in the left (red line) and right (blue line) eyes. The long infra-slow component is indicated on both diagrams by the dotted curves that were mathematically fitted to the maximal and minimal values of the pupil area. The period (TISO_long) of the long ISO is indicated on both diagrams. The short infra-slow component is indicated on both diagrams by solid lines that were mathematically fitted to the fragments of pupil size oscillations observed during ECoG activation and deactivation (see below). The period (TISO_short) of the short ISO is indicated on both diagrams. Bottom: two diagrams showing cyclic alternations of the brain state. The black diagram shows the raw ECoG signal, and the colour surface plot below indicates the dominant frequency in the ECoG signal. The hottest (red) and coldest (blue) colours represent the highest and lowest values in the power spectrum at the particular frequency calculated for the window sliding over the raw ECoG signal (step 10 s; width 300 s). Activation phases, indicated by shaded areas, are characterised by dominant theta frequency (≈4 Hz) in the ECoG signal, and deactivation phases are characterised by dominant delta frequency (≈1 Hz) in the ECoG.
Figure 3.
Graph showing the average period ± SEM (n = 13 for each value; ○) for the long ISO and short ISO components of oscillations observed in the size of the pupil of the left (red filled circles) and right (blue filled circles) eyes.
The ECoG phase (i.e., deactivation and activation) is indicated by captions at the bottom of the graph if the value was determined for this phase only. Note the lack of statistically significant differences (p = 0.73; Student’s t-tests for paired data) between the values determined for long ISO and between the values determined for short ISO.
Figure 4.
Graph showing the minimal (▿), maximal (▵) and peak-to-peak amplitude (⋄) average values ± SEM (n = 13 for each value) for infra-slow oscillations observed in the size of the pupil of the left (red filled symbols) and right (blue filled symbols) eyes.
The ECoG phase (i.e., deactivation and activation) when the values were determined is indicated by captions at the bottom of the graph. Differences marked by asterisks are statistically significant (p<0.01; Student’s t-tests for paired data).
Figure 5.
An example of the synchronisation of infra-slow oscillations of pupil areas observed simultaneously in both eyes.
C: graph showing overlaid oscillations simultaneously observed in the pupil area of the left (red line) and right (blue line) eyes. A and B: graphs showing the magnified fragments of the recording shown in graph C. D and E: colour surface plots showing the cross-correlograms calculated with a sliding window (step, 10 s; width, 300 s) for two fragments of the recording shown in graph C. The y-axis denotes the time lag of the correlation function; the x-axis denotes the time range; and the cross-correlation value is coded by colour (colour scale is given on the inset). F and G: time-averaged cross-correlograms computed from D and E, respectively; the SDs of the calculated means are shown in violet. Note that the pupil area oscillations are synchronised between the eyes during ECoG activation (A, D, F) and deactivation (B, E, G).
Figure 6.
An example of the eyeball movements resulting in shifts in the positions of the pupil centres.
Top: diagrams showing the shifts in the positions of the pupil centres of the left (red circles) and right (blue circles) eyes over the full ECoG activation-deactivation cycle. The zero of the nasal-temporal and dorsal-ventral axis was arbitrarily set at the most nasal position of the pupil centre. Grey arrows indicate the direction of the eyeball movements during cyclic transitions between ECoG deactivation (temporal positions) and activation (nasal positions). Solid lines with numbers indicate the distance between the most extreme pupil centre positions in the nasal-temporal direction. Bottom: frames of the left and right eyes taken when the pupil centre was located at the most temporal (upper images) and most nasal (lower images) positions. The graphs from the top of the figure have been overlaid over the frames to visualise the pathway of the pupil centres. The scale bar on all frames is 1 mm.
Figure 7.
An illustration of a typical correlation of eyeball movements with brain-state changes.
Top: diagrams showing the pupil centre movements in the nasal-temporal and dorsal-ventral directions, separately for the left (upper diagram) and right (lower diagram) eyes from the example in Figure 6. Bottom: two diagrams showing cyclic alternations of the brain state. The black diagram shows row ECoG signal, and the colour surface plot below indicates the dominant frequency in ECoG signal. The hottest (red) and coldest (blue) colours represent the highest and lowest values in the power spectrum at the particular frequency calculated for the window sliding over the raw ECoG signal (step 10 s; width 300 s). Activation phase, indicated by shaded area, is characterised by dominant theta frequency (≈4 Hz) in ECoG signal, and deactivation phase is characterised by dominant delta frequency (≈1 Hz) in ECoG. Note that the relocation of the pupil from the temporal to nasal position, concomitant with the transition from ECoG deactivation to activation, is markedly faster than the gradual shift of the pupil centre from the nasal to temporal position.
Figure 8.
Diagram showing the functional model of the generation of infra-slow oscillations in the pupil size of urethane-anaesthetised rats.
A detailed description of the model is given in the Results section. Note that green dashed arrows indicate plausibly polysynaptic route. Abbreviations: AAS, ascending activating system; OPN, olivary pretectal nucleus; IML, intermediolateral cell column; SCG, superior cervical ganglion; EW, Edinger-Westphal nucleus; CG, ciliary ganglion; ISO, infra-slow oscillation.