1
LETTER
Accumulation of Neural Activity in the Posterior Insula
Encodes the Passage of Time
Marc Wittmann
1,3
, Alan N. Simmons
1,3
, Jennifer L. Aron
2
, Martin P. Paulus
1,3
1
Department of Psychiatry, University of California San Diego, La Jolla, CA 92093,
USA.
2
Department of Neurosciences, University of California San Diego, La Jolla,
CA 92093, USA.
3
Veterans Affairs San Diego Healthcare System, San Diego, CA
92161, USA.
2
The experience of time, i.e. the estimation of duration, is fundamental for
perception and behavior and, therefore, essential for the survival of the
individual organism
1-3
. Over the last decades, neuroimaging,
neurophysiological and clinical neuropsychological studies have pointed to
many different brain areas involved in the processing of time
4-8
. However, the
core neural substrates and the processes accounting for the encoding of
duration, which could form a timekeeping mechanism (essentially, a `neural
clock'), are still unknown. Here we present evidence of neurophysiological
activity in circumscribed areas of the human brain that is involved in the
encoding of duration. Time-activity curves of neural activation derived from
event-related functional magnetic resonance imaging (fMRI) during a time
estimation task show that bilateral posterior insula as well as superior
temporal and inferior parietal cortices build up activation when individuals are
presented with 9 or 18 seconds tone intervals. Since the build up of neuronal
activation peaks at the end of the interval, it appears that this accumulator-
type activity encodes duration. Because of the close connection between
posterior insula and ascending internal body signals
9,10
, the accumulation of
physiological changes in body states might constitute our experience of time.
These results could be the starting point for a neural model of human time
perception in the multiple-seconds range in which specific brain regions
accumulate brain activity for the representation of duration.
To date, there is still no conclusive answer to the question of which areas of
the brain and what kind of neurophysiological processes account for the experience
of duration in humans. A number of different brain areas have been implicated in a
3
postulated time keeping mechanism: notably, the cerebellum
5
, the right posterior
parietal cortex
4
, the right prefrontal cortex
8
, and fronto- (SMA) striatal circuits
7,11-13
.
The involvement of multiple brain areas in the perception of time may be due to
different temporal-processing components that are not necessarily related to the
encoding of duration, e.g. attention, working memory and decision-making
6,14
. Neural
processes across different brain areas also depend on the duration of the judged
intervals: specifically, millisecond timing is governed by different processes than time
perception in the seconds or multiple-seconds range
3, 8, 15, 16
.
However, there is also no consensus as to what mechanisms account for our
sense of time. The most prominent models over the last years have been variants of
a pacemaker-accumulator clock where an oscillator produces a series of pulses and
the number of pulses recorded over a given time span represents experienced
duration
17-19
. Other theoretical models assume specific neuronal system properties
for encoding time not related to a pacemaker
20-22
, or propose that memory decay
processes are involved in time perception
23
.
Functional magnetic resonance imaging (fMRI) studies have revealed that
several brain areas engage during various time perception tasks which employ time
intervals up to a few seconds
8
. Due to the temporal constraints of the method (with
data acquisition times typically near 2 seconds) the detection of neurophysiological
changes over time is not possible for these short time intervals. Here we present the
missing link for a neural theory of time perception for multiple-second intervals: we
show empirical evidence of neurophysiological activity in specific areas of the brain
(recorded with fMRI) related to the encoding of durations up to 18 s. Fourteen human
subjects were instructed to reproduce tone intervals with 3, 9, and 18 s duration
4
during fMRI (Fig. 1). Each trial consisted of two consecutive phases: the encoding
and the reproduction phases, respectively. In the encoding phase, participants
listened to a 1.2 kHz tone. After a short pause, the reproduction phase was started
and consisted of the presentation of a 2 kHz tone. In this phase, participants had to
stop the presentation of this second tone when they estimated that it had reached
the length of the first tone. In the control task, subjects had to listen to a 2 kHz tone
with durations of around 3, 9, and 18 s and to press the button as quickly as possible
as soon as the tone stopped (the control phase). For details concerning the
methods, see the Supplementary Information.
In accordance with former studies employing the temporal reproduction
method
24,25
, the mean of the reproduced intervals were accurate for the 3 s standard
interval (mean reproduction: 2918 ms, S.D.: 628) and progressively shortened with
increasing interval lengths: 7576 ms (S.D. = 1434) for the 9 s interval and 12702 ms
(S.D. = 2723) for the 18 s interval (see Fig. S 1).
First, we asked which brain areas are activated during the encoding and
reproduction phase for the three different durations. Two-way (task, duration)
analyses of variance (ANOVA) were conducted to test the differences of activation
for the two contrasts: (1) encoding vs. control phase and (2) reproduction vs. control
phase (p < 0.01, corrected). Several brain areas were significantly activated for the
contrasts shown separately for the three durations (Supplementary Tables 1, 2).
Specifically, the supplementary motor area (SMA) was activated across all durations
in the encoding phase. Specifically for the 9 and 18 s intervals, in addition to other
areas, activations in posterior insular and superior temporal cortices were found.
Relative to the encoding phase, the reproduction of time intervals engaged more
5
anterior portions of the brain, which is most evident in the anterior shift of medial
frontal (partly SMA) activation during the 9 and 18 s intervals (see Fig. 2).
Additionally, in the encoding phase only posterior structures of the insular cortex
showed significant activation differences, whereas in the reproduction phase
activation in the posterior as well as (more pronounced) the anterior insula was
observed together with inferior frontal activation.
Our second question focused on the time course of neural activity in the
identified brain regions of interest (ROI) for the 9 and 18 s intervals. We found two
fundamentally different temporal activation patterns (Fig. 2 A, B): (1) a rapid-onset,
steady increase of activation over time or `climbing activity' that peaks at the end of
the stimulus (the assumed peak of the hemodynamic function = stimulus length + ca.
6 s delay) and (2) an inverted u-shape activation that increases with a considerable
delay after stimulus onset and decreases before the end of the stimulus. A factor
analysis conducted over the time courses of activation in the ROI confirms the
existence of two different factors corresponding to the two observable temporal
activation patterns (see Supplementary Table 3). In the 9 s condition, climbing
neuronal activity is detectable in the ROI that encompasses the right-sided posterior
insula and portions of the superior temporal and pre-central gyrus (Fig. 2 A). In the
18 s condition, climbing activity is visible for the ROI in the right posterior insula (and
parts of the post-central gyrus), the left posterior insula, and a right superior temporal
and inferior parietal region (Fig. 2 B). Inverted u-shape functions in the 9 and 18 s
conditions are seen for the right pre- central gyrus, the SMA bilaterally (Fig. 2 A, B)
and a right pre-post central area. A ROI encompassing the left posterior insula, parts
of the pre-post central gyrus as well as superior temporal gyrus displays
6
characteristics of both types of time activity curves, namely the observed activation
increases immediately after stimulus onset and plateaus before dropping shortly
before the end of the stimulus (Fig. 2 A).
In the reproduction phase, time activity in different ROI exhibit a similar
temporal profile, i.e. show a monotonic rise followed by a sudden drop about two to
four seconds before the actual button press (Fig. 2 C, D). This pattern can be seen
during the reproduction of the 9 s intervals for the ROI identified to be located in the
right SMA (Fig. 2 C), the right post-central, inferior parietal cortices, the left inferior
frontal cortex, the right anterior insula, inferior frontal cortex, the right inferior frontal
cortex, the left posterior insula. During the reproduction of the 18 s interval this
activation pattern emerges only ~10 s before the button press in the ROI in a right
medial frontal area, the left anterior insula, and the right anterior insula (Fig. 2 D).
Thus, while activity during the encoding phase continues until the termination of the
stimulus (presumably to adequately represent its duration), the reproduction phase
activity in the ROI peaks 2 to 4 seconds before the motor response (presumably
reflecting the moment at which the decision is made as to when to stop the tone).
The time activity curves in the encoding phase are not unlike those recently
reported in neurophysiological animal studies showing that specific climbing
neuronal activity, interpretable as representing a temporal integrator-like function,
encodes short durations
26, 27
. Neuron ensembles in pre-motor and motor cortex
28
as
well as posterior parietal cortex
29
of rhesus monkeys monotonically increase (or
decrease) their activity throughout delays up to a few seconds before a timed motor
response is made. We similarly interpret the bilateral climbing neuronal activity in
posterior insula, superior temporal and right inferior parietal gyrus as indication of an
7
ensemble accumulator process, i.e. the increasing engagement of multiple local
neural circuits to encode duration of auditory signals (thus, also the engagement of
the superior temporal cortex). Theoretical models have been developed showing
how signal accumulation over time can function as a time keeper
13,21,27
. In contrast,
the u-shape function might be involved in attention and working memory processes
supportive of the timing task.
The insular cortex is part of the extended limbic system, and is strongly
involved in subjective feeling states and interoceptive (within the body) awareness.
The posterior insula is specifically implicated as the basic receptive area for visceral
input, that is, for physiological states of the body
9,10
. Activation of the insular cortex
has repeatedly been shown during neuroimaging tasks on time perception but its
significance has seldom been discussed
6,8,14
. It is only very recently that a
conceptual framework for an anatomical and structural model of insular cortex has
been formulated. Craig
30
suggests that insular cortex integrates interoception and
the processing of emotional moments with the perception of time. In line with this
proposal we show that the posterior insula is a key neural substrate for the encoding
of duration of multiple seconds and that, consequently, the accumulation of
physiological changes in body states registered in the posterior insula may contribute
to our perception of time.
This is the first study in humans showing an integrator-like neuronal function
over time involved in the representation of duration. The finding that neural activity
accumulates in the posterior insula provides key evidence for piecing together a
theory in which interoception might function as the prime source for our subjective
experience of duration of multiple seconds.
8
METHODS SUMMARY
Functional MRI data was collected while subjects temporally reproduced tone
intervals with varying durations of 3 sec, 9 sec, and 18 sec. Each trial started with a
1.2 kHz tone presented for one of the three durations (i.e. the encoding phase). After
a short pause a 2 kHz tone (i.e. reproduction phase) was presented (see Fig. 1).
This second tone had to be stopped by the subject when she believed that it had
reached the length of the first tone. In order to contrast both the encoding and the
reproduction phases with the control task phase (see below) subjects had to also
press a button as fast as possible at the end of the encoding phase of the timing
task. This was implemented in order to have comparable attention and motor
preparation demands in all three task phases.
In the control reaction time task subjects listened to 2 kHz tones with variable
durations (control phase) and to press the button as quickly as possible as soon as
the tone stopped. Unbeknownst to the subject, the tone durations were identical to
the reproduced durations in the reproduction task of a previous behavioral session
outside the scanner. To prevent subjects from counting, a secondary memory task
was employed in both the temporal reproduction and the control task (Fig. 1; see
also the Supplementary Material).
Functional ROI were identified for the contrasts between the encoding and the
control phase as well as the reproduction and the control phase (p < 0.01,
corrected). The time activity curves from these regions were extracted for each
participant and averaged over the time points of acquisition (every 2s). Individual
time activity curves during the 9 and 18 s encoding phases were normalized (set to
zero) at the onset of the standard stimulus. Individual time activity curves for the 9
9
and 18 s reproduction phases were aligned to the actual individual reproduction
times of the participants (stopping the second tone plus the projected delay of the
hemodynamic function).
Full Methods and any associated references are available.
10
1.
Buhusi,C.V. & Meck,W.H. What makes us tick? Functional and neural
mechanisms of interval timing. Nat Rev Neurosci 6, 755-765 (2005).
2.
Wittmann,M. & Paulus,M.P. Decision making, impulsivity and time
perception. Trends in Cognitive Sciences 12, 7-12 (2008).
3.
Pöppel,E. A hierarchical model of temporal perception. Trends in
Cognitive Sciences 1, 56-61 (1997).
4.
Bueti,D., Walsh,V., Frith,C., & Rees,G. Different Brain Circuits Underlie
Motor and Perceptual Representations of Temporal Intervals. Journal of
Cognitive Neuroscience 20, 204-214 (2008).
5.
Ivry,R.B. & Spencer,R.M. The neural representation of time. Curr Opin
Neurobiol 14, 225-232 (2004).
6.
Rao,S.M., Mayer,A.R., & Harrington,D.L. The evolution of brain
activation during temporal processing. Nature Neuroscience 4, 317-323
(2001).
7.
Coull,J.T., Vidal,F., Nazarian,B., & Macar,F. Functional Anatomy of the
Attentional Modulation of Time Estimation. Science 303, 1506-1508 (2004).
8.
Lewis,P.A. & Miall,R.C. Distinct systems for automatic and cognitively
controlled time measurement: evidence from neuroimaging. Current Opinion
in Neurobiology 13, 250-255 (2003).
9.
Critchley,H.D., Wiens,S., Rotshtein,P., Öhman,A., & Dolan,R.J. Neural
systems supporting interoceptive awareness. Nature Neuroscience 7, 189-
195 (2004).
10.
Craig,A.D. How do you feel? Interoception: the sense of the
physiological condition of the body. Nat Rev Neurosci 3, 655-666 (2002).
11.
Pouthas,V. et al. Neural network involved in time perception: an fMRI
study comparing long and short interval estimation. Hum Brain Mapp 25, 433-
441 (2005).
12.
Hinton,S.C. & Meck,W.H. Frontal-striatal circuitry activated by human
peak-interval timing in the supra-seconds range. Brain Res Cogn Brain Res
21, 171-182 (2004).
13.
Jech,R., Dusek,P., Wackermann,J., & Vymazal,J. Cumulative blood
oxygenation-level-dependent signal changes support the 'time accumulator'
hypothesis. Neuroreport 16, 1467-1471 (2005).
14.
Livesey,A.C., Wall,M.B., & Smith,A.T. Time perception: Manipulation of
task difficulty dissociates clock functions from other cognitive demands.
Neuropsychologia 45, 321-331 (2007).
11
15.
Wittmann,M. Time perception and temporal processing levels of the
brain. Chronobiol. Int. 16, 17-32 (1999).
16.
Wittmann,M. et al. Effects of psilocybin on time perception and
temporal control of behaviour in humans. J. Psychopharmacol. 21, 50-64
(2007).
17.
Zakay,D. & Block,R.A. Temporal Cognition. Current Directions in
Psychological Science 6, 12-16 (1997).
18.
Church,R.M. Properties of the internal clock in Timing and time
perception (eds. Gibbon,J. & Allan,L.) 566-582 (New York Academy of
Sciences, New York, 1984).
19.
Treisman,M., Faulkner,A., Naish,P.L.N., & Brogan,D. The internal
clock: Evidence for a temporal oscillator underlying time perception with some
estimates of its characteristic frequency. Perception 19, 705-743 (1990).
20.
Karmarkar,U.R. & Buonomano,D.V. Timing in the Absence of Clocks:
Encoding Time in Neural Network States. Neuron 53, 427-438 (2007).
21.
Wackermann,J. & Ehm,W. The dual klepsydra model of internal time
representation and time reproduction. Journal of Theoretical Biology 239, 482-
493 (2006).
22.
Matell,M.S. & Meck,W.H. Cortico-striatal circuits and interval timing:
coincidence detection of oscillatory processes. Cognitive Brain Research 21,
139-170 (2004).
23.
Staddon,J.E.R. Interval timing: memory, not a clock. Trends in
Cognitive Sciences 9, 312-314 (2005).
24.
Ulbrich,P., Churan,J., Fink,M., & Wittmann,M. Temporal reproduction:
further evidence for two processes. Acta Psychologica 125, 51-65 (2007).
25.
Sawyer,T.F., Meyers,P.J., & Huser,S.J. Contrasting task demands alter
the perceived duration of brief time intervals. Perception & Psychophysics 56,
649-657 (1994).
26.
Durstewitz,D. Neural representation of interval time. NeuroReport 15,
745 (2004).
27.
Reutimann,J., Yakovlev,V., Fusi,S., & Senn,W. Climbing neuronal
activity as an event-based cortical representation of time. J Neurosci 24,
3295-3303 (2004).
28.
Lebedev,M.A., O'Doherty,J.E., & Nicolelis,M.A.L. Decoding of
Temporal Intervals From Cortical Ensemble Activity. Journal of
Neurophysiology 99, 166 (2008).
12
29.
Leon,M.I. & Shadlen,M.N. Representation of Time by Neurons in the
Posterior Parietal Cortex of the Macaque. Neuron
38, 317-327 (2003).
30.
Craig,A.D. Interoception and emotion: a neuroanatomical perspective
in Handbook of Emotion (eds. Lewis,M., Haviland-Jones,J.M. & Barrett,L.F.)
272-288 (Guilford, New York, 2008).
Supplementary Information is available.
Acknowledgments Thanks are due to Jan Churan for his invaluable help in shaping
the WinVis for Matlab scripts and to Virginie van Wassenhove for comments on the
manuscript. This work is supported by a grant from NIDA (1R03DA020687-01A1 to
M.P. and M.W.) and by a grant from the Kavli Institute for Brain and Mind (KIBM 07-
33 to M.P. and M.W.).
Author contributions M.W., A.N.S., and M.P.P. were responsible for the overall
study design. M.W. and J.L.A conducted the behavioral and fMRI experiments.
M.W., A.N.S, and M.P.P. analyzed the fMRI data and M.W. wrote the paper with the
help of all authors.
Author Information Correspondence and requests for materials should be
addressed to M.W. (wittmann@ucsd.edu).
13
Figure 1 | Task design: Trial events in the temporal reproduction and the control
reaction time task (for details, see Supplementary Information). To discourage
subjects from counting (which they were instructed not to), in both the timing and the
control task, a secondary memory task had to be performed. In the timing task,
subjects first saw for three seconds four numbers on the screen. Then, a continuous
1200 Hz tone was presented for one of three durations (3, 9, 18 seconds). After the
tone had stopped subjects had to press a button as fast as possible. After a short
pause a continuous 2000 Hz tone was presented that had to be stopped by pressing
a button when the subjects thought that it has lasted as long as the first stimulus.
Then one single number appeared on the screen and subjects had to decide by
pressing one of two buttons whether it was one of the four numbers seen at the
beginning of the trial. The control reaction time task was characterized by subjects
reacting as fast as possible with a button press when a 1200 Hz tone stopped.
Figure 2 | Brain activity related to the encoding > control (A, B) and reproduction >
control contrasts (C, D) (P < 0.01, corrected) are coded by yellow to red voxels
(activation) and blue (deactivation) and superimposed on the average of anatomical
images of the 14 subjects (sagittal and axial planes). Individual time activity curves
during the 9 and 18 s encoding phases were normalized (set to zero) at the onset of
the standard stimulus. In the encoding phase climbing brain activity can be discerned
that peaks at the end of the stimulus duration (with a delay of ca. 6 seconds
reflecting the hemodynamic response function) in the right posterior insula (R p Ins)
(9 and 18 sec, A, B, respectively) as well as in the left posterior insula (L p Ins) and
the right superior temporal (ST) and inferior parietal (IP) cortex. An inverted u-shape
14
function is detected in medial frontal (SMA). In the reproduction phase time activity
curves in the region of interest (ROI) peaks two to four seconds before the button
press (C, D). Among other regions, medial frontal as well as bilateral anterior insula
(a Ins) and the inferior frontal (IF) regions show this temporal envelope.
0
0.15
0.3
0.45
0
2
4
6
8
10 12 14 16 18 20 22
duration [sec]
ROI ac
ti
vat
i
o
n
-0.15
0
0.15
0
2
4
6
8
10 12 14 16 18 20 22
duration [sec]
R
O
I
a
c
ti
v
a
ti
o
n
0
0.15
0.3
0.45
0
2
4
6
8
10 12 14 16 18 20 22
duration [sec]
ROI activation
p Ins
p Ins
ST
ST
x = 8
z = 14
R
L
SMA
L R SMA
L p Ins, ST
R p Ins, ST
A
Encoding phase 9 s
tone duration
* projected peak of hemodynamic response
*
-0.15
0
0.15
0.3
0.45
0
2
4
6
8
10
12
14 16
18
20 22
24
26 28
30
duration [sec]
R
O
I
ac
t
i
vat
i
o
n
x = 8
z = 14
p Ins
p Ins
ST, IP
R
L
-0.15
0
0.15
0
2
4
6
8
10
12
14
16
18
20
22
24
26
28
30
duration [sec]
R
O
I a
c
ti
v
a
ti
o
n
0
0.15
0.3
0.45
0.6
0
2
4
6
8
10
12
14
16
18
20
22
24
26
28
30
duration [sec]
R
O
I
act
i
vat
i
o
n
-0.1
0.05
0.2
0.35
0
2
4
6
8
10
12
14
16
18
20
22
24
26
28
30
duration [sec]
R
O
I
act
i
vat
i
o
n
SMA
L R SMA
L p Ins
R p Ins
R ST, IP
B
Encoding phase 18 s
*
* projected peak of hemodynamic response
tone duration
z = 2
x = 4
a Ins
a Ins
IF
R SMA
IF
L
R
0
0.15
0.3
0.45
-10
-8
-6
-4
-2
0
duration [sec]
R
O
I
a
c
t
i
vat
i
o
n
-0.15
0
0.15
0.3
0.45
-10
-8
-6
-4
-2
0
duration [sec]
R
O
I a
c
ti
v
a
tio
n
-0.15
0
0.15
0.3
0.45
-10
-8
-6
-4
-2
0
duration [sec]
RO
I
act
ivati
o
n
button press
R SMA
R a Ins, IF
L a Ins, IF
C
Reproduction phase 9 s
-0.05
0.05
0.15
-18
-16
-14
-12
-10
-8
-6
-4
-2
0
duration [sec]
R
O
I a
c
tiv
a
tio
n
-0.3
-0.15
0
0.15
0.3
-18
-16
-14
-12
-10
-8
-6
-4
-2
0
duration [sec]
R
O
I
ac
t
i
vat
i
o
n
-0.1
0
0.1
-18
-16
-14
-12
-10
-8
-6
-4
-2
0
duration [sec]
R
O
I
act
i
vat
i
o
n
x = 4
z = 2
a Ins
a Ins
IF
R
L
medial F
button press
R medial F
L a Ins
R a Ins, IF
D
Reproduction phase 18 s
