The Hemodynamic Impulse Response to a Single Neural Event

Abstract

Summary: This article investigates the relation between stimulus-evoked neural activity and cerebral hemodynamics. Specifically, the hypothesis is tested that hemodynamic responses can be modeled as a linear convolution of experimentally obtained measures of neural activity with a suitable hemodynamic impulse response function. To obtain a range of neural and hemodynamic responses, rat whisker pad was stimulated using brief (Յ2 seconds) electrical stimuli consisting of single pulses (0.3 millisecond, 1.2 mA) combined both at different frequencies and in a paired-pulse design. Hemodynamic responses were measured using concurrent optical imaging spectroscopy and laser Doppler flowmetry, whereas neural responses were assessed through current source density analysis of multielectrode recordings from a single barrel. General linear modeling was used to deconvolve the hemodynamic impulse response to a single "neural event" from the hemodynamic and neural responses to stimulation. The model provided an excellent fit to the empirical data. The implications of these results for modeling schemes and for physiologic systems coupling neural and hemodynamic activity are discussed.

Long Duration Stimuli and Nonlinearities in the Neural–Haemodynamic Coupling

“…For example, Ances et al (2000) found that, taking into account neural nonlinearities, responses to brief stimuli (o4 secs) were linear, while responses to longer stimuli were better explained by a nonlinear model. Similarly, our previous work (Martindale et al, 2003) found the response to brief stimuli to be well modelled by a linear convolution of measured neural activity with a haemodynamic impulse response. An attempt to explain the apparent contradiction between human and animal studies can be found in the discussion.…”

“…The mean impulse response derived from the 1 to 4 secs stimuli was compared with the flow impulse response previously identified in Martindale et al (2003) (equivalent to equation (2) with d ¼ 2.37 secs, t ¼ 0.76 secs, and C ¼ 10.6). There was a high degree of agreement between the two impulse response estimates (r 2 ¼ 0.96, data not shown), indicating that the responses to the 1 to 4 secs stimuli could be well modelled by the previously identified LCM.…”

“…To test the generality of the model, data from a previous study which were well explained by the LCM (see Martindale et al, 2003) were re-analysed using the modified model. An acceptable fit was found under all stimulus conditions, with r 2 40.91 in all cases (data not shown).…”

“…Also shown is the impulse response function previously identified for the LCM for 2-sec 1 to 5 Hz stimuli. The term 'general' impulse response is used to denote IR 1 of equation (3) (convolved with neural activity), the 'brief' impulse response is taken to refer to IR 2 (convolved with exponentially weighted neural activity) and the 'linear' impulse response refers to the impulse response previously identified in Martindale et al (2003). Interestingly, the 'linear' and 'brief' impulse responses are very similar both in amplitude and in time course, while the 'general' impulse response is broader, with a longer latency and smaller amplitude.…”

Comprehensive correlation between neuronal activity and spin-echo blood oxygenation level-dependent signals in the rat somatosensory cortex evoked by short electrical stimulations at various frequencies and currents

“…Although most of the human fMRI studies employ the event-related task design that has a short stimulus duration, animal studies generally investigate the relationship between BOLD signals and evoked potentials with a stimulus duration of several tens of seconds (Brinker et al, 1999;Van Camp et al, 2006;Goloshevsky et al, 2007;Huttunen et al, 2008;Sanganahalli et al, 2009). Unlike BOLD studies, numerous optical studies performed using laser Doppler flowmetry (LDF) (Ngai et al, 1999;Matsuura and Kanno, 2001;Ureshi et al, 2004Ureshi et al, , 2005 and optical imaging (Devor et al, 2003;Martindale et al, 2003;Sheth et al, 2003Sheth et al, , 2004Jones et al, 2004;Hewson-Stoate et al, 2005;Martin et al, 2006) have investigated the hemodynamic responses such as CBF and deoxygenated hemoglobin changes to neuronal activity in the rat cortex with a short stimulus duration (2-5 s) (Matsuura and Kanno, 2001;Martindale et al, 2003;Sheth et al, 2003Sheth et al, , 2004Jones et al, 2004;Ureshi et al, 2004Ureshi et al, , 2005Hewson-Stoate et al, 2005;Martin et al, 2006). Thus, an animal fMRI study with short stimulus duration is expected to bridge the results of the optical studies using short stimulus durations and those of the BOLD studies using long stimulus durations.…”