Cumulative blood oxygenation-level-dependent signalchanges supportthe‘time accumulator’ hypothesis Robert Jech, 1,CA Petr Dus› ek, 1 Jir› |¤ Wackermann 3 and Josef Vymazal 2 1 Department of Neurology,1st Medical Faculty of Charles University, Kater› inska¤ 30,120 00 Prague; 2 Na Homolce Hospital, Roentgenova 37/2, 150 00 Prague,Czech Republic; 3 Institute for Frontier Areas of Psychology,Wilhelmstr. 3a, 79098 Freiburg i.Br.,Germany CA Corresponding Author: panther@tremor.anet.cz Received 29 May 2005; accepted15 June 2005 We studied time-related changes in the blood oxygenation-level- dependent signal during a time reproduction task. Nine healthy studyparticipantsretainedandreproducedstimuliofvaryingdura- tions in the multi-second range. During the encoding phase of the task, activity in the left dorsolateral prefrontal cortex inversely correlated with the interval duration, while an adjacent region in the dorsolateral prefrontal cortex showed positive correlation with duration in the reproduction phase. Cumulative signal increase during the reproduction phase, as found in the primary motor and supplementary motor areas, may also re£ect the time-sensitive behavior. Signal accumulation in the right caudate nucleus is in agreement with presumed role of basal ganglia in time perception. These results support the ‘time accumulator’ hypothesis. NeuroReport 16:1467^1471  c 2005 Lippincott Williams & Wilkins. Keywords: Prefrontal cortex;Time perception;Time reproduction; Working memory INTRODUCTION Many theories of time perception assume an ‘internal clock’ consisting of a pulse generator (pacemaker) and pulse counter (accumulator) [1], where the accumulated pulse count serves as an internal measure of time elapsed. The time accumulator can be considered a specific type of working memory [2], in which time-related information is stored. The momentary state of the accumulator may be compared with remembered reference values, which allows a participant to reproduce temporal durations. The principle of the time accumulator is common to other models of time perception, even if they do not share the concept of a ‘ticking clock’ [3], because the spontaneously charging and/or discharging accumulator can function as a time- keeper itself [4–6]. Time perception has been studied in several imaging studies [7]. Besides the block design of the functional magnetic resonance imaging (fMRI) with alternation of the active and passive phases of the task, the event-related design [8] allowed one to visualize brain areas involved during each phase of the task. Many cortical areas involved in various aspects of time perception were described [9]; however, the results do not allow one to reliably distinguish areas truly involved in time perception from those involved in other, parallel processes, accompanying any executive task. Parametric analysis, designed to search dependence between the level of the accumulated blood oxygenation- level-dependent (BOLD) signal and the temporal duration involved, is free from disadvantages inherent to conven- tional event-related design. The parametric approach has been successfully used in a time discrimination task to study the quantitative effect of attention allocation with regard to the temporal and visual aspects of a cognitive task [10]; to identify working memory load-dependent areas of the brain [11–13]; and to detect time-sensitive areas in a simple contrast design, using long and short intervals as two conditions [14]. In our study, we used a time reproduction task, with durations varied at several levels, combined with the parametric analysis, in a systematic search of the time accumulator, that is, those cortical areas where the BOLD signal covaries with the temporal duration perceived, provided the other processes, such as attention, decision- making and behavioral response, remain constant. METHODS Nine right-handed healthy volunteers (six men, three women; mean age 28 years, range 23–49 years) participated in the study. A signed, informed consent was obtained from all participants and the study was approved by the local ethics committee. Before the experimental session, the task was explained to the study participants and practiced off the magnetic resonance apparatus. The task consisted of two phases: encoding and repro- duction. During the encoding phase, participants had to retain the duration of a presented visual stimulus – a gray square with a centered red cross. The encoding interval was followed by a constant interstimulus interval (10 s), during which an indifferent stimulus (gray cross) was displayed. In the reproduction phase, participants had to reproduce the BRAIN IMAGING NEUROREPORT 0959-4965  c Lippincott Williams & Wilkins Vol 16 No 13 8 September 2005 1467 Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.