Minimum intensity of stimulus at which participants can identify its presence 50 of the time

Orthonasal Psychophysical Tools

Orthonasal tests can be broadly divided into those that test odor threshold and those that test suprathreshold olfactory function.

Odor threshold is the minimum concentration at which an odor can be detected. The corresponding operational definition relates to the concentration where an odor is detectable to a participant during 50% of trials (i.e., they will smell this stimulus at this concentration during half of the trial presentations). Odor threshold tests assess quantitative olfactory impairment—they do not require the participant to be familiar with or identify the stimulus. Generally, they use a forced choice paradigm whereby participants are asked to choose which of a set of presented stimuli contained an odorant, compared to blank, odorless distractors. Therefore, some degree of short-term working memory is required for the successful completion of the task; however, this test has a lower cognitive burden than others, which may require the use of episodic or semantic-type memory to a greater degree.276

Suprathreshold olfactory tests assess a participant's ability to evaluate odors based on their quality. They use odors of sufficient concentration such that they should be detectable to an unimpaired person (suprathreshold), and, therefore, aim to assess qualitative rather than quantitative dysfunction. Suprathreshold tools can test for odor discrimination (the nonverbal ability to differentiate between different smells) and odor identification (the ability to name an odor). Unprompted odor identification is not easy—participants rarely achieve more than a 50% correct identification rates.277 Therefore, such tests generally use verbal and/or visual cues.278 Odor discrimination and identification have a higher cognitive burden than threshold: performance in these suprathreshold tasks correlates significantly with tests of semantic memory and executive function.276 Furthermore, as odor identification performance is dependent on previous odor learning/experience, such tests are culturally specific and are influenced by age. Accordingly, odor identification tests must be validated in regional, age-matched populations.279 Testing children as young as 5 years old may be possible using adult psychophysical tests. Where this is not possible, pediatric psychophysical tests have been developed (seeTable 36.4).

The use of multicomponent psychophysical testing (i.e., combining threshold ± discrimination ± identification) is recommended by the PPOD guidelines. The reason for this recommendation is twofold. First, the diagnostic sensitivity of psychophysical tools increases when multiple subcomponents are tested. For example, in a large study of 2178 patients, the use of individual odor threshold (T), discrimination (D), or identification (I) (TDI) scores to diagnose dysfunction was less sensitive than using composite TDI scores (64%, 56%, and 47% of TDI for T, D, and I, respectively). Diagnostic sensitivity increased when subcomponent pairs were used (e.g., T + I, T + D, and I + D), but still fell short of the composite TDI score.280 Second, it is thought that odor threshold may best reflect disorders of the peripheral olfactory system (at the level of the OE), while the suprathreshold tests of discrimination and identification preferentially reflect disorders of central olfactory processing. While this interpretation must be viewed with caution, it is supported by behavioral studies showing correlation between discrimination/identification but not threshold with measures of cognition,276 lesions studies showing impaired identification but not threshold in focal cerebral excision,243 and analysis of subtest result patterns in patients with olfactory dysfunction of various causes.192 In the latter study, subtest results from 1,226 patients with hyposmia were analyzed: patients with sinonasal olfactory dysfunction were found to have particularly impaired odor threshold scores, while those with PD were particularly impaired in suprathreshold tasks. The PPOD guidelines, therefore, recommend that, ideally, multicomponent psychophysical testing be undertaken, with threshold + at least one suprathreshold olfactory test; however, any enhanced diagnostic information gathered from multicomponent testing must be offset by the increased duration of testing.

Abstract

Psychophysics is the premier research tool for studying the relationship between the physical world and its sensory representations. The first section of the article delineates the typical components of a psychophysical experiment. The second section describes the different ways in which psychophysical experiments can be classified: Class A versus Class B; Type 1 versus Type 2; performance versus appearance; forced‐choice versus non-forced‐choice; criterion‐dependent versus criterion‐free; detection versus discrimination; and threshold versus suprathreshold. The final section outlines the principal varieties of psychophysical procedure.

Pain Psychophysics: Role of Gender, Age, Race, and Ethnic Identity

A growing literature demonstrates gender differences in pain evoked by heat, cold, pressure, chemical, and electrical stimulation (Fillingim et al 2009), and the general topic of gender differences is addressed inChapter 15 by Greenspan and Traub.

Sixty years ago an anthropologist described ethnic differences in pain expression, with Jewish and Italian Mediterranean cultures being more expressive than American and Irish cultures (Zborowski 1952). This report further differentiated these groups: Italians were supposedly more present centered, whereas Jews were concerned about the future. Irish were influenced by negative social connotations of pain expression, whereas Americans were thought to genuinely be stoical. Within 2 decades these differences were partly confirmed in the experimental pain laboratory. Turksy and Sternbach, using electrical stimulation of the skin, compared the pain sensitivity of housewives in these ethnic groups who had immigrated to the United States. Both psychophysical and psychophysiological measures provided experimental confirmation of Zborowski’s observations (Sternbach and Tursky 1965,Tursky and Sternbach 1967).

Ethnic differences have been demonstrated repeatedly by using a variety of experimental pain measures. In the United States, the majority of these studies have compared Caucasians and African Americans. Beginning with the original study of Chapmen and Jones (1944), which actually preceded Zborowski’s anthropological studies, experiments have found similar or increased sensitivity in African American subjects (Edwards et al 2001;Campbell et al 2005;Mechlin et al 2005;Rahim-Williams et al 2007;Campbell et al 2008a, 2008b). In comparison to Caucasian subjects, these studies have generally found similar pain thresholds but less tolerance and increased pain ratings in response to cold pain, heat pain, and ischemic pain in African Americans. This pattern of reduced tolerance and increased sensitivity to suprathreshold stimulation has been interpreted in terms ranging from psychological mechanisms such as hypervigilance (Campbell et al 2005) to physiological mechanisms of impaired endogenous pain regulatory systems (Mechlin et al 2005).

Increased pain sensitivity has also been observed in minority groups, such as Asian Indian Singaporeans in comparison to Chinese and Malays (Tan et al 2008), South Indians in comparison to Danish Caucasians (Gazerani and Arendt-Nielsen 2005), Middle Eastern subjects in comparison to Swedes (Dawson and List 2009), and Chinese in comparison to European Canadians (Hsieh et al 2010). The study of Singaporeans was interesting because it used a natural, acute painful stimulus, cesarean section, instead of laboratory stimulation, and the dependent measures included both pain ratings and morphine consumption. A study of Libyans in Libya noted that sensitivity was decreased in the majority ethnic group (Tashani et al 2010) but still found increased sensitivity when compared with the results of a reference group of “Western” subjects from London who participated in a separate experiment (Keogh et al 2005).

1 Introduction

Psychophysics is a scientific approach to the measurement of mental processes. The empirical laws of psychophysics are based on such measures as the magnitudes of stimuli just noticeably different from each other, the proportion of errors made when comparing two stimuli, the time required to complete such judgments, and self-reports of the confidence in comparative judgments or the strength of a stimulus. Some laws define relations between physical events and their mental representations such as sensations and feelings. The theories of psychophysics provide explanations of these empirical relations and suggest neural mechanisms that give rise to experimental results.

Visual Psychophysics

Vision loss is the most common clinical presentation in XLRS patients. Visual acuity may deteriorate during the first and second decades of life, presenting as young as 3 months,98 then remain relatively stable with very slow progression of macular atrophy until the fifth or sixth decade,125 with eventual progression to legal blindness (acuity <20/200) by the sixth or seventh decade. XLRS patients are often detected when having reading difficulties and poor vision at school age. Visual acuity ranges from 20/20 to less than 20/200. The average visual acuity in young adults is around 20/70. Retinal detachment and vitreous hemorrhage may be the cause of a precipitous drop in visual acuity. Defective color vision (red–green dyschromatopsia) can also be present in XLRS patients.126 The visual field shows an absolute scotoma in the field corresponding to the location of the peripheral retinoschisis.

3 Psychophysical Methods

Psychophysics is the systematic study of sensory capacities by determining behavioral responses to physical changes in sensory stimuli. Although the theory and techniques of psychophysics were developed to investigate human sensory functions, “animal psychophysics” was subsequently established as a powerful tool for studying the sensory systems of animal subjects (see Stebbins, 1990, for a historic review). Psychophysical procedures usually require the use of operant learning tasks, but the topic is discussed here because the purpose of the task is to assess sensory function instead of learning or memory.

After subjects have initially been trained to perform an operant response such as a lever press for food or water reinforcement, a stimulus control element is introduced into the task. Animals may be asked whether they detect the presence of a sensory stimulus, or they may be required to discriminate between stimuli that vary along some chosen dimension. The simplest way to carry out such training is to use a single-response (“go/no-go”) task in which, for example, lever pressing is reinforced in the presence but not the absence of the stimulus. Evans (1982) mentions certain disadvantages to this approach and points out that forced-choice methods are generally preferable, although such methods are admittedly more labor intensive and require more elaborate and costly equipment. The major advantages of forced-choice methods are that reinforcement is available on each trial and that response topography is the same for each subject choice (i.e., press one lever when you detect the stimulus and press the other lever when the stimulus is not detected). Several options are also available with respect to stimulus presentation procedures. Readers interested in more information on this topic are referred to the review by Rice (1994).

8.2 Background

Psychophysics deals with the nature of the quantitative relationship between physical and mental qualities. Today, the practice of psychophysics is ubiquitous in all fields of neuroscience that involve the study of behaving organisms, be they man or beast. Curiously enough, the origins of systematic psychophysics can be traced to a single individual: Gustav Theodor Fechner (1801–1887). Fechner’s biography exhibits many telling idiosyncrasies. Born the son of a pastor, he studied medicine at the University of Leipzig, but never practiced it after receiving his degree. Mostly by virtue of translating chemistry and physics textbooks from French into German, he was appointed professor of physics at the University of Leipzig. In the course of studying afterimages by gazing into the sun for extended periods of time—himself being the primary and sole research participant—he almost lost his eyesight and went into deep depression in the early 1840s. This episode lasted for nearly a decade, a time which Fechner spent mostly within a darkened room. Emerging from this secluded state, he was overwhelmed by the sheer brilliant radiance of his surroundings, giving rise to his panpsychist worldview: he was now utterly certain that all things have souls, including inanimate objects such as plants and stones. Determined to share his insights with the rest of humanity, he soon started publishing on the topic, formulating an “identity theory” stating that the physical world and the spiritual world are not separate entities, but actually the same—the apparent differences resulting from different perspectives (first versus third person) onto the same object. In his view, this reconciles the incompatible dominant philosophical worldviews of the 19th century: idealism and materialism. However, his philosophical treatises on subjects such as the soul-life of plants or the transcendence of materialism were poorly received by the scientific community of the day (Fechner, 1848). In order to convince his colleagues of the validity of his philosophical notions, he set out to devise methods that would allow him to empirically link physical and spiritual realms (Fechner, 1851). His rationale being that if it can be shown that mental and physical qualities are in a clear functional relationship, this would lend credence to the notion that they are actually metaphysically identical.

Publishing the results of empirical studies on the topic in his Foundations of Psychophysics in 1860, he showed that this is the case for several mental domains, such as the relationship between physical mass and the perception of heaviness or weight. Fechner formulated several methods to arrive at these results that are still in use today. Importantly, he expressed the results of his investigations in mathematical, functional terms. This allowed for the theoretical interpretation of his findings. Doing so, he introduced notions such as sensory thresholds quantitatively.

Ironically, inventing psychophysics did not help Fechner in convincing his philosophical adversaries of the merits of his identity theory. Few philosophers of the day renounced their idealistic or materialistic positions in favor of identity theory. Most of them simply chose to ignore Fechner, while the others mostly attacked him. Consequently, Fechner spent much of the remainder of his life fighting these real or imaginary adversaries, publishing two follow-up volumes in 1877 and 1882, chiefly focusing on the increasingly bitter struggle against the philosophical establishment of Imperial Germany. Ultimately, these efforts had little tangible or lasting impact. Meanwhile, the first experimental psychologists, particularly the group around Wundt, pragmatically used these very same methods to create a psychology that was both experimental and empirical. It is not too bold to claim that they never stopped and that contemporary psychophysics derives in an unbroken line from these very roots.

The key to visual psychophysics (and psychophysics more generally) is to elicit relatively simple mental phenomena that lend themselves to quantification by presenting physical stimuli that are easily described by just a few parameters such as luminance, contrast, or spatial frequency.

It is imperative that the experimenter has complete control over these parameters. In other words, the visual stimuli that s/he is presenting have to be precise. One way to create these stimuli is to use commercially available graphics editors, most prominently Photoshop®. While this practice is very common, it comes at a cost. For example, the images created by Photoshop have to be imported by the experimental control software. It is more elegant to create the stimuli in the same environment in which they are used. More importantly, the experimenter surrenders some degree of control over the created stimuli, when using commercial graphics editors, because the proprietary algorithms to perform certain image functions are not always completely documented or disclosed. This problem is equally avoided by creating the stimuli in a controlled way within MATLAB.

28.2 Signal Detection Theory and Psychophysics

Psychophysics is the subfield of psychology devoted to the study of physical stimuli and their interaction with sensory systems. Psychophysical tasks have been extensively used to draw conclusions on how information is processed by the visual and other sensory systems. These tasks often resemble the one described in the previous section and use weak visual stimuli or stimuli embedded in noise. The subject's performance can then be analyzed using the signal detection theory methods introduced in the previous chapter. In this section, we present some of the additional formal framework used to describe and analyze psychophysical experiments. We start with a description of task design.

Yes–no rating experiments. Experiments like those described in §28.1 are called yes–no rating experiments. In these experiments, either one of two stimuli (s0 and s1) is randomly presented with equal probability. An observer is to report after each stimulus presentation which one of s0 or s1 was presented. The time-course of the task is illustrated in Figure 28.4A. In a typical situation s0 is “noise” and s 1 corresponds to a signal presented simultaneously with the noise (“signal plus noise”). In §28.1 the noise condition would correspond to the “blank” stimulus and the “signal plus noise” condition to the flash stimulus. The responses are denoted r=0 or 1 depending on whether “noise” or “signal plus noise” is chosen by the observer.

Two alternative forced-choice (2-AFC) experiments. A 2-alternative forced-choice experiment is one in which the subject is required to respond only after two successive stimulus presentations, as illustrated in Figure 28.4B. Both s0 and s1 are presented exactly once with equal probability in the two presentation intervals. After the second interval, the subject is asked to report in which interval s1 (“signal plus noise”) was presented. In the flash detection experiments described above, this corresponds to presenting the “blank” stimulus in one interval and the flash in the other interval and subsequently asking the subject to report in which of the two intervals the flash appeared. In this case, responses r=0 or 1 indicate the first or second interval, respectively.

Correct detection and false-alarm probabilities. In a yes–no rating experiment, the probability of correct detection, PD, is the probability of reporting the signal when it was indeed present, i.e., PD=P(r=1|s1) and the probability of false-alarm is the probability of incorrectly reporting the signal when it was absent, i.e., PFA=P(r=1|s0). The total error rate of the observer is given by averaging both types of errors by their probability of occurrence,

ε=12 PFA+12(1−PD).

The corresponding probability of correct response is PC=1−ε. In a 2-AFC experiment, the probability, PC, of correct response is defined similarly (i.e., probability of r=0 when s1 was presented in the first interval and r=1 when s1 was presented in the second interval).

Psychometric functions. When the strength of the signal is continuously varied over a range of values in a yes–no rating task, a plot of the detection probability as a function of signal strength is called a psychometric function (e.g., Figures 28.1 and 28.2). The term psychometric function is also applied to the probability of correct response in a 2-AFC task as a function of signal strength and sometimes to the same quantity in a yes–no rating task. It is usual to define from a psychometric function a detection threshold to be able to compare the responses of subjects across different conditions. Typically, detection thresholds are defined as 50% correct performance for yes–no rating experiments and 75% correct performance for 2-AFC experiments. These definitions are somewhat arbitrary and some authors define detection thresholds using different values (such as 68% correct performance for 2-AFCs).

ROC curves. For a yes–no rating experiment, the ROC curve is a plot of PD as a function of PFA for a fixed signal strength. In psychophysical experiments, ROC curves are often plotted for a signal strength equal to the psychophysical threshold. As explained above, such ROC curves fully characterize the performance of the observer for a fixed set of physical stimulus conditions.

Statistical distribution of responses. If we have access to some physiological variable such as the number of spikes fired by a neuron in response to “noise” and “signal plus noise”, the question then arises as to how that information can be used to “optimally” decide which of the two stimuli was presented. This question has been addressed in the previous chapter for the yes–no rating experiments. We address it here for the 2-AFC experiment.

Minimum error in a 2-AFC experiment. If the observer's response is not biased towards one of the two presentation intervals, the minimum error test in a 2-AFC experiment is to compare the likelihood ratio based on the outcome of the two presentations (x1,x2) and select response r =1 for the presentation interval with the highest likelihood ratio:

(28.4)lr(x1)>lr(x2)⇒r=0,l r(x1)<lr(x2 )⇒r=1

This result can be immediately derived by computing the Neyman–Pearson test corresponding to a threshold k=1, assuming that the presentations are independent (Exercise 3). Note that no threshold is needed, in contrast to the yes–no rating experiments considered previously. This can be understood intuitively from the fact that one presentation interval effectively serves as the threshold for the other one.

Area under an ROC curve. The area under the ROC curve for a yes–no rating task equals the expected ideal observer performance in the corresponding 2-AFC task, i.e.,

(28.5)PC=∫01PD(PFA)dPFA

(Exercise 4). The area under the ROC curve in a yes–no experiment is thus often used as a measure of discrimination performance, since it is independent of the chosen threshold and since it predicts performance in the corresponding 2-AFC task, under the assumption that the observer treats both intervals identically.

2-AFC Gaussian model. The correct response probability for the Gaussian noise model is given by

(28.6)

(Exercise 5). Comparing this equation with the equivalent one for the yes–no rating task (Eq. (27.6)) shows that the parameter dYN characterizing correct detection in the yes–no rating task is related to the equivalent parameter d2AFC by d2AFC=2dYN.

3 Attention and Perception

The psychophysics literature provides good accounts of how visual thresholds correlate with attentional investment. However, improvement in “visual acuity” is not synonymous with altered thresholds for detection, better performance, or faster reaction times. For example, acuity requires the resolution of detail, whereas detection thresholds and reaction time can involve the summation of luminance, and this might obscure detail.

There is evidence that attention improves performance in spatial resolution tasks. Cognitive scientists draw a distinction between how attention may be useful for simple detection of events and how performance can improve at those events. Although performance may improve on increased attentional investment, there has been great controversy over what orienting attention to a sensory (e.g., visual) stimulus actually does. There is general agreement that the attended stimulus receives priority, so that reaction time to it is usually faster. For example, in the visual modality, there is evidence of enhancement of brain electrical activity over extrastriate visual areas by approximately 90 milliseconds after visual presentation. On the other hand, attention is not a panacea to perception, and there is a great deal that attention cannot do. For example, it is clear that attention to a peripheral stimulus does not compensate for the lack of acuity that would be present for a foveal stimulus. Stimuli in the fovea always have an advantage in detail, although the priority for processing the input has been placed elsewhere. Thus, while orienting to a location, attention can give priority to that location (i.e., targets that appear there will be perceived more rapidly and with lower thresholds), but it cannot substitute for the acuity provided by the fovea. Although the fovea is critical for acuity, the costs in reaction time for an unexpected foveal stimulus are just as great as those for an unexpected peripheral event. Thus, visual attention influences priority or processing preference.

Whereas investing attention is frequently associated with looking directly at the scene of interest, covert attention is the ability to select visual information at a cued location, without eye movements, and to grant such information priority in processing. Researchers have shown that the performance improvement at attended locations results, to some extent, from an enhanced spatial resolution at the cued location. Findings from further psychophysical studies support the hypothesis that attention increases resolution at the attended location. Studies exploring the relationship between visual attention and contrast sensitivity show that covert attention not only improves discriminability in a wide variety of visual tasks but also can speed up the rate at which information is processed. There are findings indicating that a person’s contrast sensitivity is greater in the lower visual meridian (not the higher one). The bulk of the evidence sets limits to the effects of attention on spatial resolution and specifies that certain visual (not attentional) constraints determine aspects of spatial resolution.

2.1 Semiorders

In psychophysics one has to deal with the phenomenon of subliminal changes in intensity of physical stimuli. One stimulus may be (physically) slightly more intense than another, but a subject does not perceive the difference. For example, one might imagine a coffee-tester confronted with an extended series of cups of coffee, each with one more grain of sugar than its immediate predecessor. He/she will probably not notice the difference in taste between successive cups, but surely the difference between the first and the last cup will be clearly noticeable. In other words, indifference with respect to ≤ is not transitive. Similar observations can be made in decision theory and other branches of psychology. One way of describing such facts is to introduce sensory thresholds, thereby creating a scaling problem where one wants to scale both the subjective intensity and the magnitude of the threshold. Luce (1956) considered two conditions which seem plausible in this context (P is the set of stimuli and x≤y denotes the observation x is less intense than y):

(S1)Ifx⩽y andu⩽v,thenx⩽vor u⩽y.(S2)Ifx⩽y⩽zandw∈P,thenw⩽zonx⩽w.

One easily shows that (P,≤) is a quasiorder, i.e., without losing generality, one can assume that it is a partial order. Condition (S1) says that it does not contain Fig. 1(a) as a suborder. Similarly, condition (S2) is equivalent to the fact that it does not contain Fig. 1(b) as a suborder.

From the viewpoint of measurement the most important property of semiorders is the possibility of a numerical representation with a constant threshold. More precisely, one can show that for a finite semiorder (P,≤) there is a function f

(2)f:P→R:such that x<yiff(x)+1<f(y).

The condition of finiteness is far too restrictive for Eqn. (2), but without further restrictions it does not obtain. The height of the threshold is arbitrary, that is, the number one in Eqn. (2) can be replaced by any other positive number (with a rescaled f). More on this kind of representation theorem can be found in Roberts (1979, Chap. 6.1) or Suppes et al. (1989, Chap. 16). Clearly, the converse of this theorem is true: a structure with a representation, Eqn. (2) is necessarily a semiorder.

Although the semiorder concept seems to have its origin in the psychological context of the possibility of measurement of subjective intensity described above, it is now a well-established notion in the mathematical theory of posets and several remarkable results are connected with it. Some of them are mentioned in the following text.