Mental chronometry
Mental chronometry is the scientific study of processing speed or reaction time on cognitive tasks to infer the content, duration, and temporal sequencing of mental operations. Reaction time is measured by the elapsed time between stimulus onset and an individual's response on elementary cognitive tasks, which are relatively simple perceptual-motor tasks typically administered in a laboratory setting. Mental chronometry is one of the core methodological paradigms of human experimental, cognitive, and differential psychology, but is also commonly analyzed in psychophysiology, cognitive neuroscience, and behavioral neuroscience to help elucidate the biological mechanisms underlying perception, attention, and decision-making in humans and other species.
Mental chronometry uses measurements of elapsed time between sensory stimulus onsets and subsequent behavioral responses to study the time course of information processing in the nervous system. Distributional characteristics of response times such as means and variance are considered useful indices of processing speed and efficiency, indicating how fast an individual can execute task-relevant mental operations. Behavioral responses are typically button presses, but eye movements, vocal responses, and other observable behaviors are often used. Reaction time is thought to be constrained by the speed of signal transmission in white matter as well as the processing efficiency of neocortical gray matter.
The use of mental chronometry in psychological research is far ranging, encompassing nomothetic models of information processing in the human auditory and visual systems, as well as differential psychology topics such as the role of individual differences in RT in human cognitive ability, aging, and a variety of clinical and psychiatric outcomes. The experimental approach to mental chronometry includes topics such as the empirical study of vocal and manual latencies, visual and auditory attention, temporal judgment and integration, language and reading, movement time and motor response, perceptual and decision time, memory, and subjective time perception. Conclusions about information processing drawn from RT are often made with consideration of task experimental design, limitations in measurement technology, and mathematical modeling.
History and early observations
The conception of human reaction to an external stimulus being mediated by a biological interface is nearly as old as the philosophical discipline of science itself. Enlightenment thinkers like René Descartes proposed that the reflexive response to pain, for example, is carried by some sort of fiber—what is recognized as part of the nervous system today—up to the brain, where it is then processed as the subjective experience of pain. However, this biological stimulus-response reflex was thought by Descartes and others as occurring instantaneously, and therefore not subject to objective measurement.The first documentation of human reaction time as a scientific variable would come several centuries later, from practical concerns that arose in the field of astronomy. In 1820, German astronomer Friedrich Bessel applied himself to the problem of accuracy in recording stellar transits, which was typically done by using the ticking of a metronome to estimate the time at which a star passed the hairline of a telescope. Bessel noticed timing discrepancies under this method between records of multiple astronomers, and sought to improve accuracy by taking these individual differences in timing into account. This led various astronomers to seek out ways to minimize these differences between individuals, which came to be known as the "personal equation" of astronomical timing. This phenomenon was explored in detail by English statistician Karl Pearson, who designed one of the first apparatuses to measure it.
Purely psychological inquiries into the nature of reaction time came about in the mid-1850s. Psychology as a quantitative, experimental science has historically been considered as principally divided into two disciplines: Experimental and differential psychology. The scientific study of mental chronometry, one of the earliest developments in scientific psychology, has taken on a microcosm of this division as early as the mid-1800s, when scientists such as Hermann von Helmholtz and Wilhelm Wundt designed reaction time tasks to attempt to measure the speed of neural transmission. Wundt, for example, conducted experiments to test whether emotional provocations affected pulse and breathing rate using a kymograph.
Sir Francis Galton is typically credited as the founder of differential psychology, which seeks to determine and explain the mental differences between individuals. He was the first to use rigorous RT tests with the express intention of determining averages and ranges of individual differences in mental and behavioral traits in humans. Galton hypothesized that differences in intelligence would be reflected in variation of sensory discrimination and speed of response to stimuli, and he built various machines to test different measures of this, including RT to visual and auditory stimuli. His tests involved a selection of over 10,000 men, women and children from the London public.
Welford notes that the historical study of human reaction times were broadly concerned with five distinct classes of research problems, some of which evolved into paradigms that are still in use today. These domains are broadly described as sensory factors, response characteristics, preparation, choice, and conscious accompaniments.
Sensory factors
Early researchers noted that varying the sensory qualities of the stimulus affected response times, wherein increasing the perceptual salience of stimuli tends to decrease reaction times. This variation can be brought about by a number of manipulations, several of which are discussed below. In general, the variation in reaction times produced by manipulating sensory factors is likely more a result of differences in peripheral mechanisms than of central processes.Strength of stimulus
One of the earliest attempts to mathematically model the effects of the sensory qualities of stimuli on reaction time duration came from the observation that increasing the intensity of a stimulus tended to produce shorter response times. For example, Henri Piéron proposed formulae to model this relationship of the general form:where represents stimulus intensity, represents a reducible time value, represents an irreducible time value, and represents a variable exponent that differs across senses and conditions. This formulation reflects the observation that reaction time will decrease as stimulus intensity increases down to the constant, which represents a theoretical lower limit below which human physiology cannot meaningfully operate.
The effects of stimulus intensity on reducing RTs was found to be relative rather than absolute in the early 1930s. One of the first observations of this phenomenon comes from the research of Carl Hovland, who demonstrated with a series of candles placed at different focal distances that the effects of stimulus intensity on RT depended on previous level of adaptation.
In addition to stimulus intensity, varying stimulus strength can also be achieved by increasing both the area and duration of the presented stimulus in an RT task. This effect was documented in early research for response times to sense of taste by varying the area over taste buds for detection of a taste stimulus, and for the size of visual stimuli as amount of area in the visual field. Similarly, increasing the duration of a stimulus available in a reaction time task was found to produce slightly faster reaction times to visual and auditory stimuli, though these effects tend to be small and are largely consequent of the sensitivity to sensory receptors.
Sensory modality
The sensory modality over which a stimulus is administered in a reaction time task is highly dependent on the afferent conduction times, state change properties, and range of sensory discrimination inherent to our different senses. For example, early researchers found that an auditory signal is able to reach central processing mechanisms within 8–10 ms, while visual stimulus tends to take around 20–40 ms. Animal senses also differ considerably in their ability to rapidly change state, with some systems able to change almost instantaneously and others much slower. For example, the vestibular system, which controls the perception of one's position in space, updates much more slowly than does the auditory system. The range of sensory discrimination of a given sense also varies considerably both within and across sensory modality. For example, Kiesow found in a reaction time task of taste that human subjects are more sensitive to the presence of salt on the tongue than of sugar, reflected in a faster RT by more than 100 ms to salt than to sugar.Response characteristics
Early studies of the effects of response characteristics on reaction times were chiefly concerned with the physiological factors that influence the speed of response. For example, Travis found in a key-pressing RT task that 75% of participants tended to incorporate the down-phase of the common tremor rate of an extended finger, which is about 8–12 tremors per second, in depressing a key in response to a stimulus. This tendency suggested that response times distributions have an inherent periodicity, and that a given RT is influenced by the point during the tremor cycle at which a response is solicited. This finding was further supported by subsequent work in the mid-1900s showing that responses were less variable when stimuli were presented near the top or bottom points of the tremor cycle.Anticipatory muscle tension is another physiological factor that early researchers found as a predictor of response times, wherein muscle tension is interpreted as an index of cortical arousal level. That is, if physiological arousal state is high upon stimulus onset, greater preexisting muscular tension facilitates faster responses; if arousal is low, weaker muscle tension predicts slower response. However, too much arousal was also found to negatively affect performance on RT tasks as a consequence of an impaired signal-to-noise ratio.
As with many sensory manipulations, such physiological response characteristics as predictors of RT operate largely outside of central processing, which differentiates these effects from those of preparation, discussed below.