Short Communication
Accuracy of respiratory symptom perception in different affective contexts

https://doi.org/10.1016/j.jpsychores.2004.12.005Get rights and content

Abstract

Objective

The accuracy of respiratory symptom perception was investigated in different affective contexts in participants (N=48) scoring high or low for negative affectivity (NA).

Methods

Within-subject correlations were calculated between two subjective ratings and their respective physiological referent (the rating of “deeper breathing” and respiratory volume, and rating of “faster breathing” and breathing frequency) across nine consecutive breathing trials. Three different air mixtures were used: room air, air enriched with 7.5% CO2, and with 10% CO2. For half the participants, the trials were framed in a pleasant context, created by adding a pleasant odour to the air mixture in addition to information announcing pleasant feelings as a result of breathing the air mixtures. The other half received the trials in a distressing context: A foul smelling odour was added and the information announced unpleasant feelings.

Results

High-NA persons were overall less accurate than were low-NA persons in the perception of respiratory volume. For breathing frequency, high-NA persons were significantly less accurate in the distressing condition than in the pleasant one, whereas for low-NA persons, the information frame did not matter.

Conclusion

The study shows that the accuracy of respiratory symptom reports is reduced in high-NA persons, especially in a distressing context.

Introduction

Physicians rely, to a large extent, on subjective symptom reports to infer underlying disease processes, often implicitly assuming a direct linear relationship between a subjective symptom report and an underlying somatic state. In other words, they assume accurate interoception. Interoception implies multiple levels of processing, from the peripheral components of the visceral and autonomic system up to the higher centers in the brain [1], [2], [3]. The latter centers allow for a modulation of interoception by perceptual-cognitive and affective processes, which may either distort or increase interoceptive accuracy. Indeed, accurate interoception requires the ability to detect, discriminate, attend to, interpret, and express these sensations correctly [4], [5]. Its assessment involves some index of the within-subject correspondence between a self-reported sensation and a clear, objective physiological referent.

Interoceptive accuracy research can be an important aid towards a better understanding of medically unexplained symptoms (MUS). MUS are commonly met, not only among somatization patients [6] and persons suffering from so-called functional syndromes [7], but also among healthy individuals [5], [8]. Moreover, a large part of the symptoms reported by patients diagnosed with a well-known disease cannot be explained by pathophysiological processes [9], [10], [11]. It is obvious that the incidence of these symptoms has costly clinical and social implications.

Interestingly, reporters of MUS are commonly marked by high negative affectivity (NA). Trait NA is a mood-related disposition to experience negative emotions. High-NA persons can be described as emotionally unstable, distressed, anxious, and highly irritable. They tend to worry and focus on the negative side of themselves, the others, the future, and the world in general [12], [13], [14], [15]. NA is believed to be quasi-identical to other concepts such as trait anxiety [16], neuroticism [12], and pessimism versus optimism [17].

A robust correlation of about .30 to .50 has been found systematically between NA ratings and subjective symptom scores. These subjective complaints are rarely associated with NA-related differences in objective health problems [12], [18], [19], nor with differences in autonomic or respiratory responses [20], [21]. In the respiratory domain, it has been observed that high-NA asthmatics report more symptoms, are more inclined to use medication, and appear to have a higher chance to become hospitalized, regardless of their pulmonary status [22], [23]. Patients with medically unexplained dyspnea report both more intense dyspnea and more anxiety than a wide variety of patients with severe pulmonary illnesses [24].

This relationship between NA and increased symptom reporting has been attributed to attentional and interpretative biases. It is assumed that high-NA individuals are hypervigilant towards internal sensations and inclined to interpret these sensations in a negative way [15]. However, both components may, in fact, predict opposite effects as regard the accuracy criterion of interoception. Intuitively, one would expect better accuracy among interoceptively vigilant persons. On the other hand, the interpretative bias may cause amplified symptom reports, suggesting a weak relationship between physiological sensations and their subjective reportings. Because typical symptom measures aggregate frequency or intensity ratings across negatively phrased items (e.g., shortness of breath), it is difficult to disentangle both biasing influences.

Empirical research on the relationship between NA and interoceptive accuracy has yielded inconsistent findings. Some results point to better accuracy among high-NA persons [25], [26], [27], while other studies suggest that high-NA individuals are actually poorer perceivers [28], [29], and still, others show no relationship [30], [31], [32], [33], [34], [35], [36]. The discrepancy among these results may, in part, be due to differences in investigated physiological responses, in research paradigms, and in populations. In addition, the role of the informational context and task instructions [37] may have been overlooked. For example, the cue competition hypothesis of Pennebaker [5] suggests that the amount of external information may affect attentional resources available to process internal (bodily) information, influencing the number and accuracy of reported symptoms.

Not only the amount but also the affective value of the available context information may be important. In the present study, we wanted to investigate the interoceptive accuracy of respiratory symptom perception in high- and low-NA persons as a function of different affective contexts. Respiratory symptoms were induced through inhalation of CO2-enriched air and were framed either in a pleasant, reassuring context or in an unpleasant, distressing context. We focused on a narrowly selected set of neutrally phrased subjective sensations with a clear physiological referent, namely, on the subjective rating of faster breathing and its correspondence with objective respiratory frequency and on the rating of deeper breathing and its correspondence with tidal volume. Neutrally phrased sensations rather than typical complaints were chosen to calculate within-subject correlations, because negatively labeled respiratory symptoms (such as shortness of breath, suffocation, etc.) would not allow to distinguish between the perceptual aspect of interoception and the possible bias towards a negative interpretation of the sensation.

Previously, we have reported preliminary evidence for lower interoceptive accuracy of respiratory symptom perception in high-NA persons. This evidence was based on a post hoc analysis of data collected for another purpose. The data allowed to calculate within-subject correlations between minute ventilation (frequency per minute×tidal volume) and the subjective item “faster and/or deeper breathing” across 10 subsequent breathing trials of 2 min [37]. Various air mixtures, involving either or not 5.5% CO2, and either a foul or fresh smelling odour, were inhaled. In addition, half of the participants were given a positive, reassuring information frame, while the other half received negative, rather distressing information about the induced sensations. In the positive information frame, there was no significant difference between high- and low-NA persons, whereas in the negative frame condition interoceptive accuracy was much lower, but only in high-NA persons.

The present study intended to replicate and extend this finding using an improved methodology. First, rather than using a subjective item, referring to two different aspects of breathing at once (“faster and/or deeper breathing”) and relate it to minute ventilation (frequency/min×tidal volume), we computed within-subject correlations between each of these components separately (respiratory frequency with “faster breathing” and tidal volume with “deeper breathing”).

Second, in the preliminary study, the set of breathing trials for a single participant contained both positively and negatively smelling odours within the same information condition. In the present study, a foul smelling odour was only used to reinforce the distressing information frame, and, conversely, a fresh smelling odour was only given in the pleasant information frame.

Third, a more balanced set of breathing trials was administered in a semirandomized order: Three trials contained room air and three contained 7.5% and three 10% CO2. In this way, a wider range of respiratory responses was induced compared with the previous study, in which only two different air mixtures (room air and 5.5% CO2 enriched air) were used. Each trial within each group of three trials was mixed with one out of three different odours with similar affective valence ratings, meaning that each of the nine trials consisted of one particular combination of a gas mixture and an odour. Different odours of the same valence were used to increase the variety and keep the participant alert across the complete set of trials.

Finally, although we focused on the within-subject correlation for respiratory frequency and tidal volume, we also collected other subjective sensations. Part of them were positively phrased, the other part was negatively phrased, to tap the effects of each information frame on the subjective symptom reports.

Consistent with our earlier findings, we hypothesized that high-NA persons would show reduced interoceptive accuracy compared with low-NA persons and that this would be more so in a distressing context.

Section snippets

Participants

Forty-eight female students (age between 18 and 23 years) were selected from the total group of first year psychology students (n>500). Half of them scored high on NA, while the other half were low. To increase the reliability of the between-group difference, participants had to score either below or above the total group median on the State-Trait Anxiety Inventory (STAI; median=41) and Positive and Negative Affect Schedule (PANAS; median=22). The STAI was used as the first criterion to select

Interoceptive accuracy

For deeper breathing, high-NA persons were overall less accurate than low-NA persons were [F(1,44)=4.91, P<.05; see Fig. 1]. For faster breathing, NA interacted with information [F(1,44)=3.95, P=.05; see Fig. 2]. Simple main effects showed that high-NA persons were less accurate in a distressing than in a pleasant situation [F(1,22)=4.94, P<.05], whereas for low-NA persons, the situation did not matter. Homogeneity of variances (Levene's test) did not significantly differ among the

Discussion

Overall, the present data largely confirm preliminary findings from a previous study [37]: Persons with high levels of NA are interoceptively less accurate, as measured by two respiratory symptoms with a clear physiological referent. The symptoms were neutrally phrased to avoid interpretative biases to distort the accuracy measure. In addition, it was shown that, for respiratory frequency, also the affective tone of the context played an important role: High-NA persons showed particularly low

Acknowledgments

The first author is a Research Assistant, supported by the Research Foundation–Flanders (FWO–Vlaanderen).

References (63)

  • C Nimnuan et al.

    Medically unexplained symptoms: an epidemiological study in seven specialties

    J Psychosom Res

    (2001)
  • R Rapee et al.

    Cognitive mediation in the affective component of spontaneous panic attacks

    J Behav Ther Exp Psychiatry

    (1986)
  • PM Salkovskis et al.

    Affective responses to hyperventilation: a test of the cognitive model of panic

    Behav Res Ther

    (1990)
  • OG Cameron

    Interoception: the inside story—a model for psychosomatic processes

    Psychosom Med

    (2001)
  • AD Craig

    How do you feel? Interoception: the sense of the physiological condition of the body

    Nat Rev Neurosci

    (2002)
  • CB Saper

    The central autonomic nervous system: conscious visceral perception and autonomic pattern generation

    Annu Rev Neurosci

    (2002)
  • H Leventhal et al.

    Affect, cognition, and symptom perception

  • JW Pennebaker

    The psychology of physical symptoms

    (1982)
  • JM Robbins et al.

    Attributions of common somatic symptoms

    Psychol Med

    (1991)
  • H Ursin et al.

    Sensitization, subjective health complaints, and sustained arousal

    Ann NY Acad Sci

    (2001)
  • JW Pennebaker et al.

    Blood pressure estimation and beliefs among normotensives and hypertensives

    Health Psychol

    (1988)
  • JW Pennebaker et al.

    The perception of general vs specific visceral activity and the regulation of health-related behavior

  • DJ Cox et al.

    Accuracy of perceiving blood glucose in IDDM

    Diabetes Care

    (1985)
  • PT Costa et al.

    Neuroticism, somatic complaints, and disease: is the bark worse than the bite?

    J Pers

    (1987)
  • D Watson et al.

    Negative affectivity. The disposition to experience aversive emotional states

    Psychol Bull

    (1984)
  • D Watson et al.

    Development and validation of brief measures of positive and negative affect: the PANAS scales

    J Pers Soc Psychol

    (1988)
  • D Watson et al.

    Health complaints, stress, and distress: exploring the central role of negative affectivity

    Psychol Rev

    (1989)
  • JA Taylor

    A personality scale of manifest anxiety

    J Abnorm Soc Psychol

    (1953)
  • MF Scheier et al.

    On the power of positive thinking: the benefits of being optimistic

    Cur Dir Psychol Sci

    (1993)
  • AJ Barsky et al.

    Cardiorespiratory symptoms in response to physiological arousal

    Psychosom Med

    (1998)
  • S Cohen et al.

    Emotional style and susceptibility to the common cold

    Psychosom Med

    (2003)
  • Cited by (98)

    • Somatic symptom disorder

      2023, Encyclopedia of Mental Health, Third Edition: Volume 1-3
    View all citing articles on Scopus
    View full text