Can we learn to sigh


1. Introduction

Do you sigh often? Do you know why? And do you do it on purpose? A sigh of a colleague can disturb us. It may be a sign of boredom or stress. But sighing can also be such a relief. Sighing is such a natural response that you do not reflect on it. But is it just a response? Or is it a behaviour that can be learned, just like any other human behaviour?

Sighing itself is a specific sort of breathing. In this article and during the experiment, a sigh will be defined as a deep breath, with a tidal volume at least twice as large as the mean tidal volume in a surrounding representative time interval (Wilhelm, Trabert, & Roth, 2001). A sigh has an important function. Whereas the physiological functions of a sigh have been known, the psychological functions of a sigh have been focus of research only recently.

Although many people consider breathing as static and stable, the respiratory system is complex and dynamic with many feedback mechanisms (Vlemincx et al., 2013). Under the influence of homeostatic processes, the respiratory system aims for stability. This stability is important to maintain the blood CO2 and O2-levels. Within a changing environment, stability is not enough to survive. The homeostatic process has to be sensitive as well. It has to be able to adapt to changing gas levels under the influence of internal and external perturbations. Research conducted by Baldwin et al. (2004) showed that there are important fluctuations in breathing, that consist, however, of correlations between successive breaths, both on the short term (short-range respiratory memory) and on the long term (long-range respiratory memory). Other researchers have found similar results (Bruce & Daubenspeck, 1995).

Bruce and Daubenspeck (1995) pointed out that total respiratory variability consists of non-random and random variability. Non-random variability is caused by the delay in the feedback loops when the respiratory system interacts with other internal physiological systems. Non-random variability is an example of the dynamic homeostatic process of respiration. Breathing patterns are also influenced by external demands, such as emotions. These external perturbations cause noise, or random variability. Random variability is important for a flexible and adaptive response to environmental demands. Too much randomness can be noticed by an increase in total variability (random) and the loss of the short-range memory or non-random variability (Baldwin et al., 2004).

Vlemincx, Van Diest, Lehrer, Aubert and Van den Bergh (2010) showed that regulating respiratory variability is an important function of a sigh. The experiment of Vlemincx, Van Diest, et al. (2010) identified that, due to an increase in correlated variability, variation changes to more non-random variability after a sigh. The change in variation suggests that a sigh acts as a resetter of non-random breathing variability, and therefore is an important regulator of respiratory stability and flexibility. Different mechanisms influence the occurrence of a sigh: chemical, physiological, and psychological parameters play a role in the path towards and directly after a sigh. Chemically, a lower O2 and higher CO2 level in the blood act as facilitators for a sigh to occur (Soltysik & Jelen, 2005), hereby reducing hypoxia and hypercapnia. Physiologically, sighing is important to prevent the collapsing of the lung alveoli (Bendixen, Smith & Mead, 1964). A sigh promotes lung compliance and the efficiency of gas-exchange (Antonaglia, Pascotto, De Simoni, & Zin, 2006). Patroniti et al. (2002) found that patients with acute respiratory distress syndrome (ARDS) could improve their lung mechanics by sighing more frequently. Another important characteristic of a sigh is the switch from sympathetic activity to parasympathetic activity. The importance of this switch was mostly illustrated in relation to babies with SIDS (Sudden Infant Death Syndrome; Franco et al., 2003; Galland, Taylor, Bolton & Sayers, 2000). Lastly, research shows that sighing not only affects chemical and physiological responses; emotion regulation is also an important function of sighing.

An experiment of Vlemincx, Van Diest and Van den Bergh (2015) shows that most sighs are observed during a negative low-arousal, or any high-arousal emotional state. The resetting character of a sigh can explain this, since a sigh occurs when there is less correlated variability (Vlemincx, Van Diest, Lehrer, Aubert & Van den Bergh, 2010), which is the case with negative and high-arousal emotions. Psychologically, a sigh relates to negative events such as stress, and appears more frequently with patients with panic disorder (Wilhelm, Trabert & Roth, 2001). Other research indicates a relation between sighing and relaxation, so that relaxation training often uses deep breaths to suppress dyspnoea (Hirose, 2000) and to decrease craving (McClernon, Westman & Rose, 2004). Relief is one of the emotions that have been investigated in combination with sighing.



In three experiments, Vlemincx et al. (2009) examined this relation between sighing and relief. They propose that the association with relief also exists in a stressful context without any specific ending of the stressor. They argue that relief is an elicitor as well as a result of sighing. Therefore, the psychological feature of sighing is to relieve stress. If a relief effect can be seen after the occurrence of a sigh, Vlemincx et al. (2009) asked themselves if it is possible that people sigh to experience relief? This could explain the frequent sighing by patients with a panic disorder (Wilhelm et al., 2001). Wilhelm et al. (2001) found that these patients sigh more often and deeper than a control group. Moreover, the CO2 levels of patients with a panic disorder did not return to the baseline after three breaths following a sigh, as was the case in controls. This explains the hypocapnia observed with patients with a panic disorder. Wilhelm et al. (2001) argued that the hypocapnia within these panic disorder patients could be caused by sighing too often. The question arises whether panic disorder patients sigh too often because they learned from experience that the relief effect will follow a sigh.

When people sigh with no physical or chemical cause, it is not an automatic response anymore, but a behaviour. The learning theories formed within behavioural psychology may provide an answer to this typical behaviour. The operant condition theory by Skinner explains this by learning the association between a sigh and the relief effect that follows the sigh (Hermans, Eelen & Orlemans, 2007). If sighing in a stressful context leads to relief, then the sigh will make an end to a negative situation. This escaping behaviour, known as negative reinforcement, can be defined as a certain behaviour that will end a negative situation. On the other hand, it is possible that relief could act as a positive reinforcement of a sigh. Both reinforcement mechanisms would lead to the same result: the person will perform the behavior more often. The person learns that sighing leads to a relief effect on the one hand, and the end of the stressful context on the other hand. So it is possible that this person will sigh more often, just to experience relief or to escape the stressful situation. However, Greenspoon (1962) found that acquiring this relation is not always done consciously. Sometimes, discovering the relation between a behaviour and its consequence can happen unconsciously and automatically.



Learning theories could also explain the maladaptive sighing behaviour of panic disorders patients (Wilhelm et al., 2001). These patients have learned to escape a stressful situation by sighing. A sigh only helps to bring your respiratory system back to baseline when you encounter an acute stressor. A chronic stressor could mean that sighing alone is not enough to go back to the baseline-breathing pattern. When the sigh-rate is disproportionate, the short term memory will be lower, and the breathing pattern will be irregular, as can be seen in patients with a panic disorder (Wilhelm et al., 2001). The relation between sighing and conditioned responses is not yet researched.

In this experiment, our main research question is: Are sighs conditioned responses? People can learn that sighing leads to different outcomes. If participants learn that by sighing in certain conditions, they experience reward in the form of relief of a breathing resistance, whilst in other conditions, it leads to a punishment, being a heavier breathing resistance, we predict that they will sigh more often in the reward conditions. Because literature shows that panic disorder patients show a different sighing pattern, and that alexithymia and anxiety sensitivity are risk factors for panic disorder patients (Cucchi et al., 2012), we want to explore whether the conditioning of sighs is different for persons with high alexithymia or anxiety sensitivity, compared to persons with low alexithymia or anxiety sensitivity. Since panic disorder patients sigh often despite hypocapnia following a sigh, a punishment after a sigh may not lead to lower sigh frequencies in persons with high alexithymia and/or anxiety sensitivity.


2. Method

2.1 Participants

Fifteen men (Mage: 22.29; SDage: 2.31) and 29 women (Mage: 21.86; SDage: 3.11) volunteered to participate in exchange for course credits or a compensation of 15 euro. They all reported to be in good health and free of any respiratory and cardiovascular diseases. The Ethics Committee of the Faculty of Psychology and Educational Sciences and the Faculty of Medical Sciences approved the experiment.


2.2 Measurements

2.2.1 Physiological measures.

The NeXus-10 MK II® (Mind Media B.V.) was used to collect all physiological data. Two electrical wires sewn into two elastic bands, to assess rate and volume of breathing, were placed around the ribcage and abdomen. Two pre-gelled electrodes, measuring electromyography (EMG), were placed on the forehead to detect muscle tone of the medial Frontalis. Together with three pre-gelled electrodes to measure electrocardiography (ECG), two sensors to measure skin conductance at the medial phalanges of the index and middle finger, and one control electrode on the C1 vertebra were connected to the NeXus. The NeXus sent the physiological data to BioTrace software for real-time feedback and data storage. Capnography (POET II, Criticare) was used to continuously monitor pCO2 from the exhaling tube.

2.2.2 Self-report measures (Appendix C).

A health questionnaire was used to determine the health status of the participant. Specifically the questions about heart diseases and respiratory problems were important for the rest of the experiment. The validated Dutch translation of the Anxiety Sensitivity Index – 3 (ASI: Taylor et al., 2007; De Jong, 2008) and of the Toronto Alexithymia Scale (TAS: Bagby, Parker & Taylor, 1994) were used to determine the level of respectively anxiety sensitivity and alexithymia. Bagby, Parker and Taylor (1989) define alexithymia as a construct with four essential parts: “(a) difficulty in identifying and describing feelings, (b) difficulty in distinguishing between feelings and bodily sensations, (c) restrictive imaginative processes and (d) a cognitive style that is concrete and reality-based.”. A self-report scale (Appendix A) was used to measure the perceived dyspnoea during the experiment. Whenever the participant felt a shortness of breath, they were asked to indicate this on a scale from zero (No perceived dyspnoea) to one hundred (Maximally imaginable dyspnoea).


2.3 Procedure

The participants were individually invited to participate in the study ‘Catch Your Breath’, a study on the influence of breathing resistance on different physiological and psychological parameters.

Upon arrival, participants were asked to complete the informed consent (Appendix B), and to fill in the health questionnaires, the ASI and the TAS on the computer. For the preparation of the physiological measurements, the participant was asked to wash their left hand with cold water, to have a better skin conductance signal. Their forehead was scrubbed, again to have a better signal. Then the experimenter placed the five electrodes on the participant and connected these with the NeXus. After checking the quality of the physiological data, the experimenter gave the specific instructions of the experiment (Appendix D). The participant was informed about the structure and goal of the experiment and what they could expect. They were told to indicate their feeling of dyspnoea using a self-report rating scale, of which the labels were introduced to them. The participant was asked not to talk or move, except to turn the rating scale to indicate the experienced dyspnoea. The participant was also reminded that he could stop whenever he wanted. After the instructions, the breathing mask was installed on the mouth and nose of the participant. The breathing mask was attached to a two-way valve, connected to one tube for inhaling and one for exhaling. This was necessary to induce breathing resistance only when the participant inhaled, so the exhalation process was without resistance. The participant was told before putting on the mask that the face mask and the tubes could cause some resistance as well, and the best way to deal with this would be to just let his breathing spontaneously adjust. After the installation of the breathing mask, the instructions were shortly repeated and the experiment began.


2.4 Design

The experiment (Appendix E) consisted of four blocks, two acquisition blocks and two test blocks. Each block consisted of 16 trials. During the first phase of each trial, subjects were exposed to a breathing resistance for 40 seconds, the dyspnoea phase - cued for each subject by either a circle or a triangle. During the second phase of each trial, the breathing resistance was removed, leading to relief of dyspnoea - cued for each subject by the other symbol, i.e., a triangle or a circle. In the first two blocks, the acquisition blocks, sighing during the relief of dyspnoea phase was monitored online. Two different reinforcement symbols (a star or a square) were presented during the relief of dyspnoea phase, predicting whether a sigh during this phase would be rewarded by the prolonging of the relief phase with 20 seconds, or would be punished by the immediate switch to breathing resistance for 20 seconds after the sigh. Following the reward or punishment phase, the next trial started. One acquisition block consisted of reward symbols, while the other one consisted of punishment symbols. If no sigh occurred in the relief of dyspnoea phase, this phase lasted 20 seconds.



The order of the blocks and the symbols were counterbalanced between subjects. After the acquisition blocks, the same trials were repeated in two test blocks. Again, participants were exposed to 40 seconds breathing resistance (dyspnoea phase), which was followed by a 20 seconds period during which the resistance was removed (relief of dyspnoea phase) and the reinforcement symbol for the respective subject (either the star or the square) was shown. However, these sighs were not rewarded or punished. The next trial followed immediately after the relief of dyspnoea phase. The experiment was finished after the fourth block and the participant was thanked for his/her participation.


3. Data analysis & Results

The physiological signals and stimulus presentation events were synchronized using Matlab R2015a (The MathWorks). The respiratory signals were visualized, screened for artefacts and pre-processed using VivoSense software (Vivonoetics). First, a qualitative diagnostic calibration was performed to integrate ribcage and abdominal traces and calculate the contribution of both to a total respiratory volume. Next, respiratory time and volume of each breath was determined. Sighs were determined as deep breaths: breaths with a respiratory volume at least twice as large as the mean respiratory volume in each block (Wilhelm et al., 2001). Only sigh rates, rating scale responses and the questionnaires data will be reported here. All the results were analysed within subjects. A repeated measures ANOVA was used for these analyses (Appendix F). Participants with missing values of one or more variables were excluded in the analyses of these specific variables. This exclusion could lead to varying degrees of freedom in some tests. Statistica 64 (Dell Inc., 2015) was used to statistically analyse the data.

The following questions were investigated by the described analyses, and were significant at α < .05.


3.1 Manipulation check: Were sighs accurately monitored online during the acquisition phase?

In order to check whether sighs were detected online in a correct way and thus to check whether sighs were consistently rewarded or punished, the sigh frequency during the relief of dyspnoea phase was compared between trials that were rewarded or punished vs. trials that were not. A repeated measures ANOVA was performed with on one hand the sigh frequency during the relief of dyspnoea phase as dependent variable, and on the other hand the sigh detection - and thus reinforcement (Yes or No) - as a first independent variable, and type of reinforcement (Reward or Punishment) as a second independent variable. The dependent variable sigh frequency can be defined as the mean sum of sighs in each phase of a trial. This unit of measure will be used during the whole experiment.

There was no significant interaction between the type of reinforcement (Punishment or Reward) and the reinforcement of a sigh (Yes or No) for sigh frequency in the relief of dyspnoea phase (F (1, 31) = 1.56, p = .22). There was however a significant main effect of reinforcement on sigh frequency (F (1, 31) = 86.13, p = .000). The sigh frequency was significantly higher when the experimenter detected a sigh online, and thus reinforced (punished or rewarded) a sigh (Figure 1).


Afbeelding verwijderd. Afbeelding verwijderd.

Figure 1. Sigh frequency during the relief of dyspnoea of the acquisition phase for the reinforcement of a sigh by type of reinforcement. The vertical lines denote the standard error.


3.2 Main research question: Are sighs conditioned responses?

To examine whether sighs are conditioned responses, we compared sigh frequencies in response to relief of dyspnoea in the test phase while presenting a cue that was coupled with reward in the acquisition phase, with sigh frequencies in response to relief of dyspnoea in the test phase while presenting a cue that was coupled with punishment during the acquisition phase. A repeated measures ANOVA was performed with sigh frequency during relief of dyspnoea in the test phase as dependent variable and the type of reinforcement (Reward vs. Punishment) as independent variable. We found that the sigh frequency in the relief of dyspnoea of the test phase was significantly higher during the reward condition in comparison to the punishment condition (F (1, 43) = 5.69, p = .022 (Figure 2)).

Afbeelding verwijderd.

Figure 2. Sigh frequency during relief of dyspnoea of the test phase for the type of reinforcement (Reward and Punishment). The vertical lines denote the standard error.


3.3 Exploratory question 1: Is the conditioning of sighs dependent on anxiety sensitivity?

To examine if anxiety sensitivity influences the conditioning of sighs, we compared sigh frequency between the two types of reinforcement during the relief of dyspnoea of the test phase for high and low scores on ASI. A median split was used to divide the participants in a low (X ≤ 31; N = 21) and high anxiety sensitivity group (X > 31; N = 23). A repeated measures ANOVA was performed with sigh frequency as dependent variable and the type of reinforcement and the anxiety sensitivity as independent variables.

There was no interaction between the type of reinforcement and a high or low ASI score (Figure 3). Although there was a significantly higher sigh frequency in the reward condition (F (1, 42) = 5.45, p = .024), people with a high ASI score did not sigh significantly more than people with a low ASI score (F (1, 42) = 2.07, p = .16).

Afbeelding verwijderd.

Figure 3. Sigh frequency during relief of dyspnoea of the test phase for the variable type of reinforcement (Reward and Punishment) by ASI-score. The vertical lines denote the standard error.


3.4 Exploratory question 2: Is the conditioning of sighs dependent on alexithymia?

In order to check if alexithymia has an influence on the conditioning of sighs, we compared the sigh frequency between the two types of reinforcement (Reward and Punishment) over different levels of alexithymia during the relief of dyspnoea of the test phase. The TAS divides the participants in three groups: No alexithymia (X ≤ 51; N=30); possible alexithymia (52 ≤ X ≤ 60; N=10); and alexithymia (X ≥ 61; N=4) (Bagby, Parker & Taylor, 1994). The same distribution was used in the repeated measures ANOVA, to evaluate the interaction and effects of the independent variables (type of reinforcement and alexithymia) on sigh frequency.

No significant interaction was found between TAS and the type of reinforcement for sigh frequency (F (2, 41) = 1.38, p = .26). There was also no significant effect of the TAS score on the sigh frequency (F (2, 41) = .33, p = .72).

Afbeelding verwijderd.  

Figure 4. Sigh frequency during relief of dyspnoea in test phase for the two conditions of the variable reinforcement (Reward and Punishment) by the TAS score. Vertical bars denote the 0.95 confidence interval.


3.5 Exploratory question 3: Does learned sighing reduce subjective dyspnoea?

To investigate if conditioning of sighs influences the subjective dyspnoea feeling, we compared the subjective feeling of dyspnoea as dependent variable in the reward condition with the subjective feeling during the punishment trials during the test blocks. A repeated measures ANOVA was performed with subjective dyspnoea during the dyspnoea phase and the relief of dyspnoea phase as dependent variables, and type of reinforcement (reward or punishment) as independent variable.

During the dyspnoea of the test phase, there is no significant difference between the reward and punishment conditions (F (1, 43) = .04, p = .83). This effect is also not significant during the relief of dyspnoea phase (F (1, 43) = .05, p = .82).

Afbeelding verwijderd.

Figure 5. Subjective feeling of breathlessness during the relief of dyspnoea of the test phase for the two conditions of the variable type of reinforcement (Reward vs. Punishment). The vertical lines denote the standard error.




4. Discussion

We conducted an experiment to investigate whether participants sigh more during cues that were previously coupled with a reward, consisting of relief of dyspnoea, compared to cues that were previously coupled with a punishment, consisting of a breathing resistance. Based on the literature review, we also explored if behavioural traits and personality, such as anxiety sensitivity and alexithymia, could influence the conditioning of sighing. We used anxiety sensitivity and alexithymia because they have been linked with panic disorder patients (Cucchi et al., 2012), who have been found to have different sighing patterns (Wilhelm et al., 2001). Lastly, in order to know whether the applied breathing resistance was experienced as such in contrast with the relief of breathing resistance, their subjective feeling of dyspnoea was investigated.

We hypothesized that participants would sigh more when a reward would be expected, than when a punishment would be expected. The experiment confirmed that participants sighed more during the reward condition than during the punishment condition in the test trials. We could say that people learned to sigh, due to the expected reward or punishment. As manipulation check, we investigated if the experimenter monitored the sighs correctly during the relief of dyspnoea in the acquisition phase. The significant higher sigh rate of the participants in the conditions where the experimenter detected a sigh, suggests that the sighs were correctly identified and rewarded or punished. Besides the main research question, we had several exploratory questions.

First, we found no significant evidence that people with a higher anxiety sensitivity responded differently to the conditioning of a sigh. Secondly, alexithymia had no significant influence on the conditioning of sighs either. Lastly, we explored whether the subjective feeling of breathlessness was lower during conditions in which sighs were increased due to conditioning. This was not the case in this study.

The study confirms that sighing can act as a conditioned response. Following the principles of the behavioural therapy (Hermans, Eelen & Orlemans, 2007), operant conditioning states that a behaviour will be repeated if followed by a positive reinforcement or a negative reinforcement. Vlemincx, Van Diest and Van den Bergh (submitted) found that a sigh was followed by relief. The positive reinforcement of a sigh through relief, as well as the negative reinforcement through the avoidance of a stressful situation, could explain the contingency between sighing and relief, found in several studies (Soltysik & Jelen, 2005; Vlemincx, Taelman, Van Diest, & Van den Bergh, 2010; Vlemincx et al., 2009). Relief would act as a natural (or intrinsical) reward. Moreover, the finding that sighing can be a learned behaviour could also explain several experiments where sighing related to negative emotions (Vlemincx, Taelman, De Peuter, Van Diest, & Van Den Bergh, 2011; Vlemincx et al., 2015). If relief is a natural positive reinforcement of sighing, the participant can learn that sighing could lead to feelings of relief when overwhelmed by negative emotions. This learning process could operate both consciously and unconsciously, as noticed in the experiment of Greenspoon (1962).

The finding that people learn to sigh can explain the different sighing patterns found in panic disorder patients (Wilhelm et al., 2001). An important characteristic of these patients is the higher score on alexithymia and anxiety sensitivity (Cucchi et al., 2012). Therefore, we hypothesised that participants with a higher score on anxiety sensitivity and alexithymia would have a higher sigh frequency. Our results are in line with the findings that panic disorder patients sigh more. Although not significant, the participants with a higher anxiety sensitivity showed a higher sigh frequency than participants with a lower anxiety sensitivity.

More specifically, based on earlier research (Wilhelm et al., 2001), we expected that participants with a higher score on these characteristics would sigh more in the punishment condition. This happens because the carbon dioxide level with panic disorder patients does not recover following a sigh as well as in healthy controls. This would imply that panic disorder patients paradoxically create dyspnoea by sighing, as they sigh often and do not restore carbon dioxide following a sigh. Nevertheless, they maintain high sigh rates. Since we did not find differences in sigh frequencies during punishment between persons with high vs. low anxiety sensitivity or alexithymia, we could not provide evidence to confirm these findings. Participants who scored higher on alexithymia sighed slightly more during the punishment conditions. Due to a lack of sufficient participants with high alexithymia, we cannot draw any conclusions on these effects. However, the direction of these non-significant effects is in line with our hypothesis, and worth exploring in the future.

Recent findings may suggest why the reinforcement effects of relief in response to sighing in panic disorder patients may be higher than in controls. Participants with a high anxiety sensitivity experienced, in addition to psychological relief (as did participants with low anxiety sensitivity), also physiological relief in response to sighs. Whereas both persons with low and high anxiety sensitivity experience subjective relief following a sigh, only high anxiety sensitive persons show a decline in muscle tension in response to sighs (Vlemincx, Van Diest & Van den Bergh, submitted). This effect could function as an extra reinforcement of the sighing behaviour of high anxiety sensitive persons, and maybe panic disorder patients. Further research with panic disorder patients is required to investigate the special characteristics regarding learning effects of sighing in this group for valid statements on this topic. If there is an effect of alexithymia on learned sighing behaviour, this mechanism could explain several mechanisms in alexithymia related disorders where sighing is not a homeostatic response (Finesinger, J.E., 1943).

There are certain limitations to this study that should be taken into account. First, the study was conducted on 44 participants. This population was sufficient for the evaluation of the general research questions. However, when further dividing in smaller groups of, for example, alexithymia participants, these groups became too small to make statements about the observed effects. To increase the clinical validity, further studies should target these specific populations. Next, because participants consisted mainly of university students, generalising the results to older populations is not yet possible. A third limitation can be found in the manipulation of the breathing resistance. All the participants received the same amount of breathing resistance. Although we see that all the participants experienced the breathing resistance, the subjective feeling of the resistance was different for all the participants. In further research this could be eliminated by letting the participant choose the resistance which results in the same subjective breathlessness for all the participants at the beginning of the experiment. A fourth limitation implies that operant conditioning can proceed consiously and unconsciously (Greenspoon, 1965). The participants were not asked if they knew the purpose of the study, as a result, we could not tell if the participants consciously experienced the learning effect or that this happened unconsciously. No further statements can be made on this subject. As a fifth limitation we argue that although a learning effect was found, we can not assign the effect to a higher frequency of sighs in the reward condition, or to a lower frequency of sighs in the punishment condition. We can only say that the difference during the test phase was significant. Finally, in the punishment trials, the participant would experience an extra long breathing resistance between the sigh in the recovery phase and the recovery phase of the next trial in the acquisition. This happens because the breathing resistance of the first phase followed immediately on the punishment by breathing resistance of the third phase of the previous trial. For the participant, it was not possible to distinguish the two phases. This could mean that some of the participants did sigh during the punishment conditions in the acquisition as a homeostatic response. Further research could use another punishment method, for example one not linked with breathing.


5. Conclusion

As conclusion, we can say that sighing, besides being a natural response with an important homeostatic function (Vlemincx, Van Diest, et al., 2010), can be a conditioned behaviour as well. We investigated if participants could learn to sigh by inducing breathing resistance when they sighed during the punishment signal, and by rewarding them when they sighed during the reward signal. We concluded that they did sigh more during the reward condition than during the punishment condition. Moreover, we observed some natural sighs as well. When they had long periods of breathing resistance in de punishment condition, the particpants sighed more than expected. We did not find any significant effects of alexithymia and anxiety on the sighing behaviour. But the results show promising outcomes, if investigated with more alexithymia and high anxiety sensitivity participants. Further research is also needed to investigate these sighing learning processes with panic disorder patients.


Antonaglia, V., Pascotto, S., De Simoni, L., & Zin, W. A. (2006). Effects of a sigh on the respiratory mechanical properties in ALI patients. Journal of Clinical Monitoring and Computing, 20(4), 243–249. DOI: 10.1007/s10877-006-9028-x

Bagby, R. M., Parker, J. D. A. & Taylor, G. J. (1994). The twenty-item Toronto Alexithymia Scale-I. Item selection and cross-validation of the factor structure. Journal of Psychosomatic Research, 38, 23-32

Baldwin, D. N., Suki, B., Pillow, J. J., Roiha, H. L., Minocchieri, S., & Frey, U. (2004). Effect of sighs on breathing memory and dynamics in healthy infants. Journal of Applied Physiology (Bethesda, Md. : 1985), 97(June 2004), 1830–1839. DOI: 10.1152/japplphysiol.00298.2004

Bartlett, D. (1971). Origin and regulation of spontaneous deep breaths. Respiration Physiology, 12(2), 230–238.

Bendixen, H.H., Smith, G.M., & Mead, J. (1964). Pattern of ventilation in young adults. Journal of Applied Physiology, 19, 195-198.

Bruce, E.N., Daubenspeck, J.A., 1995. Mechanisms and analysis of ventilatory stability. In: Dempsey, J.A., Pack, A.I. (Eds.), Regulation of Breathing. Marcel Dekker, NY, pp. 285–313.

Cucchi, M., Cavadini, D., Bottelli, V., Riccia, L., Conca, V., Ronchi, P., & Smeraldi, E. (2012). Alexithymia and anxiety sensitivity in populations at high risk for panic disorder. Comprehensive Psychiatry, 53(6), 868–74. DOI: 10.1016/j.comppsych.2012.01.005

Finesinger, J.E. (1943). The spirogram in certain psychiatric disorders. Am J Psychiatry 100,159–169.

Franco, P., Verheulpen, D., Valente, F., Kelmanson, I., de Broca, A., Scaillet, S., Kahn, A. (2003). Autonomic responses to sighs in healthy infants and in victims of sudden infant death. Sleep Medicine, 4, 569–577.

Galland, B. C., Taylor, B. J., Bolton, D. P. G., & Sayers, R. M. (2000). Vasoconstriction following spontaneous sighs and head-up tilts in infants sleeping prone and supine. Early Human Development, 58(2), 119–132. DOI: 10.1016/S0378-3782(00)00070-0

Greenspoon, J. (1962). Verbal conditioning and clinical psychology. In: A.J. Bachrach (Ed.), Experimental foundations of clinical psychology, p.510-553. New York: Basic Books.

Hermans, D., Eelen, P., & Orlemans, H. (2007). Inleiding tot de Gedragstherapie (6th ed.). Houten: Bohn Stafleu van Loghum.

Hirose (2000). Restlessness or respiration as amanifestation of akathisia: Five case reports of respiratory akathisia. Journal of Clinical Psychiatry, 61, 737–741.

McClernon, F. J.,Westman, E. C., & Rose, J. E. (2004). The effects of controlled deep breathing on smoking withdrawal symptoms in de- pendent smokers. Addictive Behaviours, 29, 765–772.

Parker, J. D. A., Bagby, R. M., & Taylor, G. J. (1989). Toronto alexithymia scale, EPQ and self-report measures of somatic complaints. Personality and Individual Differences, 10(6), 599–604. DOI: 10.1016/0191-8869(89)90219-5

Patroniti, N., Foti, G., Cortinovis, B., Maggioni, E., Bigatello, L. M., Cereda, M., & Pesenti, A. (2002). Sigh improves gas exchange and lung volume in patients with acute respiratory distress syndrome undergoing pressure support ventilation. Anesthesiology, 96(4), 788–794. DOI: 10.1097/00000542-200204000-00004

Soltysik, S., & Jelen, P. (2005). In rats, sighs correlate with relief. Physiology and Behavior, 85(November 2004), 598–602. DOI: 10.1016/j.physbeh.2005.06.008

Taylor, S., Zvolensky, M.J., Cox, B.J., Deacon, B., Heimberg, R.G., Ledley, D.R., Abramowitz, J.S., Holaway, R.M., Sandin, B., Stewart, S.H., Coles, M., Eng, W., Daly, E.S., Arrindell, W.A., Bouvard, M., Cardenas, S.J. (2007). Robust dimensions of anxiety sensitivity: Development and initial validation of the Anxiety Sensitivity Index-3. Psychological Assessment, 19. Geautoriseerde vertaling: Peter J. de Jong (2008). Rijksuniversiteit Groningen.

Vlemincx, E. (2008) The psychophysiological functions of sighing (doctoral dissertation), KULeuven, België

Vlemincx, E., Abelson, J. L., Lehrer, P. M., Davenport, P. W., Van Diest, I., & Van Den Bergh, O. (2013). Respiratory variability and sighing: A psychophysiological reset model. Biological Psychology, 93(1), 24–32. DOI: 10.1016/j.biopsycho.2012.12.001

Vlemincx, E., Taelman, J., Van Diest, I., & Van den Bergh, O. (2010). Take a deep breath: The relief effect of spontaneous and instructed sighs. Physiology and Behavior, 101(1), 67–73. DOI: 10.1016/j.physbeh.2010.04.015

Vlemincx, E., Taelman, J., De Peuter, S., Van Diest, I., & Van Den Bergh, O. (2011). Sigh rate and respiratory variability during mental load and sustained attention. Psychophysiology, 48, 117–120. DOI: 10.1111/j.1469-8986.2010.01043.x

Vlemincx, E., Van Diest, I., De Peuter, S., Bresseleers, J., Bogaerts, K., Fannes, S., Van Den Bergh, O. (2009). Why do you sigh? Sigh rate during induced stress and relief. Psychophysiology, 46(5), 1005–1013. DOI: 10.1111/j.1469-8986.2009.00842.x

Vlemincx, E., Van Diest, I., Lehrer, P. M., Aubert, A. E., & Van den Bergh, O. (2010). Respiratory variability preceding and following sighs: A resetter hypothesis. Biological Psychology, 84, 82–87. DOI: 10.1016/j.biopsycho.2009.09.002

Vlemincx, E., Van Diest, I., & Van den Berg, O. (2015). Emotion, sighing, and respiratory variability. Psychophysiology, n/a–n/a. DOI: 10.1111/psyp.12396

Vlemincx, E., Van Diest, I., & Van den Berg, O. (submitted). A sigh of relief or a sigh to relieve: The psychological and physiological relief effect of sighs. Physiology & Behavior.

Wilhelm, F. H., Trabert, W., & Roth, W. T. (2001). Characteristics of sighing in panic disorder. Biological Psychiatry, 49(7), 606–614. DOI: 10.1016/S0006-3223(00)01014-3

Universiteit of Hogeschool
KU Leuven
Thesis jaar
Elke Vlemincx