While writing this paper, I often stopped to take a breather. While waiting to see if it would be accepted, I was breathless with anticipation. I hoped I wouldn't choke while presenting it. When the presentation was finished, I could breathe easy. These metaphors demonstrate the close connection between the physical act of breathing and the mental states of anxiety and its opposite-relaxation.

Anxiety isn't the only influence on breathing patterns; all emotions affect respiration. Psychologists investigating this relationship require some form of electronic patient-monitoring equipment, partially because the very act of watching one's breathing changes its pattern. Reliable and accurate respiration monitors assist psychologists in the evaluation and treatment of patients' mental and emotional states. Also, by revealing a patient's breathing patterns over a period of time, the respiration monitor can be a useful investigative tool in the diagnosis of neurotic and psychotic individuals.

New IC technologies from Maxim Integrated Products provide an accurate, low-cost solution for the design requirements of high-end applications such as respiration monitors.

A Respiration Monitor with Smart-Sensor Technology

The respiration monitor of Figure 1 displays breathing patterns, giving an approximation of the amplitude of respiration. Several important parameters used to detect anxiety are visible on this monitor: rate of breathing, regularity of the breathing pattern, and the duration of pauses after exhalation and before inhalation. Because calm, positive emotions usually produce a pattern of longer exhalation than inhalation, the ratio of inhalation to exhalation time can serve as an anxiety indicator.

Figure 1: This block diagram depicts a respiration monitor.

The monitor depicted in Figure 1 uses a silicon piezoresistive transducer (PRT) to detect the decreases and increases of pressure corresponding to inhalation and exhalation. The PRT output is fed to a signal-conditioning IC, which corrects for errors inherent in the PRT and then passes a compensated voltage signal to the analog/digital converter (ADC). The ADC output (a digitized version of the pressure signal) is then fed to a PC interface and converted to RS-232 levels. These in turn are passed to a PC, which displays the respiration waveform and allows for analysis of the parameters.

The PRT Sensor

PRTs are commonly configured as a closed Wheatstone bridge. When pressure is applied to an active-bridge PRT (Figure 2a), resistances of the diagonally opposed legs change equally and in the same direction. As the resistances of one pair of diagonally opposite legs increases with pressure, the resistances of the other pair decreases. A half-active-bridge PRT (Figure 2b) exhibits resistance changes in only half of the bridge. The advantages of full- or half-active PRT sensors include high sensitivity (>10mV/V), good linearity at a constant temperature, and the ability to track pressure changes without signal hysteresis up to the destructive limit.

Figure 2: (a) All four legs of an active-bridge PRT respond to pressure. (b) For a half-active-bridge PRT, only two legs respond to pressure.

Engineers have always employed PRTs in low- and medium-accuracy applications, but designers of higher-end applications have traditionally been forced to use strain gauges-despite the higher cost. However, new IC technologies that allow for the accurate correction of PRT sensors enable the use of these devices in high-end applications.

PRT Sensor Errors

The chief obstacle in correcting a PRT sensor is its wide range of error. The variety in PRT sensor manufacturing methods produces various types of errors and a range of error magnitudes. Even within a given model, these error magnitudes vary appreciably from one transducer to the next.

Common PRT errors include ". . .strong nonlinear dependence of the full-scale signal on temperature (up to 1%/°Kelvin), large initial offset (up to 100% of full scale or more), [and] strong drift of offset with temperature" (Konrad et al, 1999, 12).

At a given temperature, both of the PRT types in Figure 2 maintain their bridge resistances (between V_CC and ground) at a level that is fairly constant over a wide range of pressures. But as the temperature increases, the bridge resistance increases significantly. If the bridge is powered with a constant-current source, the result is an increasing bridge voltage.

PRT sensitivity increases as the bridge voltage increases with temperature. When the bridge voltage is held constant, a PRT's sensitivity to pressure decreases with the temperature. Thus, sensitivity is a function of two opposing factors: temperature and the temperature-dependent bridge voltage. These changes in bridge resistance or bridge voltage can be exploited by modern signal-conditioning ICs to correct the sensitivity errors over temperature in a PRT. ICs use the change in bridge resistance to correct for variations in sensitivity and also offset, over temperature.

A Traditional Correction Scheme

The circuit in Figure 3 compensates PRTs to a reasonable level of accuracy. It allows for the adjustment of offset, offset drift with temperature, and sensitivity drift with temperature. Related to sensitivity drift is the full-span output drift over temperature; these two parameters change proportionally in response to temperature. The relationship between offset and full-span output is shown in Figure 4.

Figure 3: A traditional correction scheme for PRTs features temperature-sensitive resistors.

The circuit's zero-trim resistors compensate the sensor's offset voltage at room temperature, and the resistors R_ts and R_tz (or R'_tz) correct for temperature errors. As described earlier, bridge resistance rises with increasing temperature, which increases the voltage across the sensor. That additional voltage increases the sensor's sensitivity, making its output voltage higher for a given pressure.

Figure 4: A PRTs offset and full-span output constitute the full-scale output.

When the voltage across the sensor is held constant, the sensor's sensitivity decreases with increasing temperature. Because the positive-going sensitivity coefficient caused by increasing bridge resistance with temperature is greater than the negative-going sensitivity coefficient, the full-span output tends to increase with temperature. Resistor R_ts negates this effect by shunting increasing amounts of the bridge current as temperature rises. In a similar manner, R_tz or R'_tz corrects the offset drift. Either R_tz or R'_tz is included in the circuit, depending on the direction of the offset drift with temperature.

The chief problem with this compensation scheme is circuit interaction among the compensation components, which makes calibration cumbersome and limits the achievable accuracy. Additionally, electronic trimming is not feasible with this technique.

A Modern Correction Scheme

In Maxim's MAX1457, a signal-conditioning IC (Figure 5) drives the respiration monitor's sensor and corrects the sensor errors. The IC contains a controlled current source that drives the sensor and an ADC that digitizes the sensor's bridge voltage. This voltage is a product of current from the current source and the temperature-dependent bridge resistance.

Figure 5: A specialized IC (MAX1457) that provides current-source excitation and compensation for the pressure sensor yields 0.1% accuracy.

Also included in the MAX1457 are a programmable-gain amplifier (PGA) that amplifies the sensor's differential output and five digital/analog converters (DACs) that correct various sensor errors. Because the sensor output is a low-level signal, the PGA output voltage is not sufficient to drive the ADC. For that reason, the MAX1457's internal op amp is used to boost the PGA output to a suitable level.

Bridge voltage increases with temperature, so temperature dependence can be used to compensate full-span temperature errors. With constant-voltage bridge excitation, the full-span output (FSO) decreases with temperature, resulting in a full-span output temperature coefficient (FSOTC) error. When the bridge voltage is made to increase with temperature at a rate that compensates for the decrease in full-span sensitivity with temperature, the FSO remains constant.

Figure 6: This circuitry within the MAX1457 compensates for offset and full-span temperature errors.

Figure 6 demonstrates how the MAX1457 implements this scheme for correcting full-span output errors due to temperature. Using the digitized bridge voltage from the ADC output, the chip determines which previously calculated correction coefficient stored in EEPROM should be applied to the FSOTC DAC. The resulting DAC output voltage then changes the current level feeding the bridge. This new current level compensates the FSO by adjusting the bridge voltage to compensate for the change in the sensor's sensitivity at a particular temperature. The chip applies the analog bridge voltage to the FSOTC DAC's reference input, providing an additional correction between each successive pair of digital numbers supplied by the ADC to the EEPROM.

The same technique compensates offset over temperature, except the OffsetTC DAC voltage is fed to a summing junction at the PGA output instead of to the MAX1457 current source.

Temperature coefficients are calculated and stored in EEPROM. In most instances, the sensor data is taken at various pressures with the sensor and MAX1457 at the lowest temperature, then the same data in taken with the sensor and MAX1457 at the highest temperature. Using this data from the temperature extremes, software written for the MAX1457 calculates the four correction coefficients (FSO, FSOTC, Offset, and OffsetTC). These four coefficients correct the PRT's first-order errors. For the accuracy level required by this respiration monitor, a fifth coefficient to correct pressure nonlinearity is considered unnecessary.

To achieve 0.1% accuracy, the MAX1457 allows compensation at specific temperatures, with recalculation of the FSOTC and the OffsetTC at each temperature. The user may select up to 120 calibration points. If sensor output errors were perfectly repeatable, the accuracy of a sensor-MAX1457 combination would be better than 0.1%.

The MAX1457 compensation technique has a significant advantage over the traditional approach shown in Figure 3. The MAX1457 eliminates interaction between compensation components by separating the offset and span adjustments; it compensates offset at the PGA, and separately adjusts the FSO via the current source. Another advantage is the extra accuracy made available by specific adjustments at various temperatures. That method is inherently more accurate than one based on external resistors, whose values cannot compensate the sensor precisely at specific temperatures.

The MAX1457 compensation technique was developed by MCA in 1994. Maxim acquired MCA in 1997, and the original MCA staff continues to develop the technology with the enhancement of Maxim's own technology and worldwide distribution network.

Simpler Compensation ICs

The 16-bit resolution of the MAX1457's correction DACs is more than is necessary for a respiration monitor. The part was chosen because it includes the extra op amp needed to boost the monitor's low-level sensor signals.

Although the MAX1457 offers greater precision than is warranted for this application, its ability to compensate for temperature errors is needed even for modest variations of temperature. A change of 10°C commonly produces a 3% change in the FSO of a PRT. Because the MAX1457 enables the respiration monitor to operate over a wide temperature range, the monitor's potential applications could include space exploration and scuba diving.

Figure 7: A MAX1450 signal conditioner operating with external laser-trimmed resistors provides 1% accuracy.

The functions performed by the MAX1450 signal conditioner (Figure 7) are similar to those of the MAX1457, but resistors rather than DACs are used to set the error correction. Because the MAX1450 uses far fewer calibration points than does the MAX1457, its accuracy is 1%. MAX1450 chips are commonly included in hybrids with laser-trimmed resistors to provide a low-cost solution.

Figure 8: A MAX1458/MAX1478 signal conditioner operating with internal 12-bit DACs provides 1% accuracy.

A third IC, the MAX1458/MAX1478 in Figure 8, provides the same basic compensation techniques as the other two ICs, but includes 12-bit compensation DACs. MAX1458/MAX1478 devices also include an EEPROM for onboard storage of the compensation coefficients. The MAX1458/MAX1478 signal conditioner provides the same 1% accuracy as the MAX1450.

MAX1450/MAX1458/MAX1478 devices compensate a sensor by calculating the four correction coefficients using pressure data measured at two temperatures-usually the extremes of the operating-temperature range. Unlike the MAX1457, these devices do not allow for additional corrections of temperature errors at user-selected temperature levels. For a more detailed discussion of these compensation schemes, refer to Konrad et al, 1999, and Dancaster et al,1997.

Respiration and Its Connection to Emotional States

The following section shows the connection between emotion and respiration, and demonstrates that the breath-pattern monitor is a useful investigative and possibly therapeutic tool. Another aim is to point out the specific aspects of the waveforms that correspond to particular emotional states.

Links between Respiration and Emotion

Anthropologists have concluded that certain basic emotions appear in every culture. Frans Boiten, a psychological researcher, lists the six basic emotions as "joy, sadness, fear, anger, surprise, and disgust" (Boiten, 1998, 31). According to one theory, each of these six emotions evokes distinct autonomic responses-including changes in respiration. Other theorists posit the idea that particular emotions per se do not specifically cause changes in breathing. Instead, emotions land somewhere along a dimension such as calm vs. excited or passive vs. active coping, and the placement along the dimension determines the breathing pattern (Boiten, 1998, 30).

Boiten's study demonstrated that both theories have merit, but only in certain situations. The breath holding that occurs when one experiences disgust or pain is best explained by the theory that specific emotions cause certain respiratory responses. The same holds true for the quick inhalation that occurs when one is startled.

On the other hand, the placement of an emotion along the negative vs. positive affect dimension explains other results. A negative emotion enhanced the variability of the breathing pattern; especially affected was the volume of air inhaled and exhaled. Decreases in the pause after exhalation occurred to some extent when a negative affect was elicited. Mild positive emotion seemed to decrease the volume of air breathed.

Other results were best interpreted by whether the subject was engaged in active or passive coping while the breathing pattern was recorded. Active coping produced a "relatively fast, shallow and regular breathing pattern. Contrarily, unavoidable aversive stimulation…has been known to engender a passive coping attitude, which is in the study reflected by a relatively deep and irregular breathing pattern. Thus we have shown that detection of and distinguishing between respiratory responses to a variety of affective and cognitive demanding tasks requires measurement of the fine structure of the respiratory cycle" (Boiten, 1998, 48, 49).

Through a comprehensive review of the literature discussing emotions and respiration, Boiten and other researchers drew the following conclusions about what emotions, behaviors, and thought processes accompany different types of breathing (Boiten et al, 1994, 119-122).

Fast and Deep Breathing
When the depth and rate of breathing increases, it is usually associated with "states of excitement. By 'excitement,' we mean the state that generates undirected, aimless behavior under conditions in which directed action is blocked or restrained. Feelings of excitement can be understood as the felt urge towards such undirected action." This form of breathing is closely associated with a fight or flight response. (Boiten et al, 1994, 119).

Fast and Shallow Breathing
"…this pattern appears to blend readiness for action with some degree of inhibitory control…. [It]…seems typical for attempts to tune in to behavioral demands that may necessitate a restrained, precise, and goal-directed (as opposed to massive) type of physical action" (Boiten et al, 1994, 120).

Slow and Deep Breathing
A relaxed, resting state typically evokes slow, deep breathing. Also an "unrestrained emotional readiness to act" may bring out this type of breathing (Boiten et al, 1994, 120, 121).

Slow and Shallow Breathing
"…primarily indicative of states that are characterized by withdrawal from the environment and by passiveness, and might be encountered during depressed and unhappy, as well as happy, but unexcited moods" (Boiten et al, 1994, 121).

Thoracic-Abdominal Balance
Thoracic dominance breathing occurs when movement of the chest dominates while in abdominal dominance breathing, abdominal movement dominates. "Thoracic dominance appears primarily to correspond to unpleasant affect, tenseness or anxiety, and abdominal dominance to pleasant emotional states or relaxation" (Boiten et al, 1994, 121).

Irregularity of Breathing
"A number of studies have shown that breathing may become irregular under conditions of emotional upset, excitement, and task involvement." Changes in the volume of air breathed, the rate of breathing, and slope of the volume of air breathed vs. time occur from breath to breath in normal breathing. "...little is known about the mechanisms that augment the occurrence of respiratory irregularities during emotional behaviour" (Boiten et al, 1994, 122).

Respiration Patterns in Normal, Neurotic, and Psychotic Individuals

Psychologists use a number of diagnostic tools to evaluate the mental and emotional status of patients. In addition to the numerous tests that can be administered, a psychologist's trained eye often picks up behavioral nuances usually missed by the layman.

One diagnostic aid is the breathing pattern of an individual. Certain features of these patterns distinguish mentally healthy people from those who are less healthy. A number of those features discovered or confirmed by one study of the breathing patterns of normal, neurotic, and psychotic individuals are listed below (Clausen, 1951).

Clausen measured the timing of respiration patterns and created a rough estimate of its amplitude using a pneumograph. He measured both the thoracic and abdominal breathing of normal, neurotic, and psychotic people for a brief period on each of five days. Five indicators of neurosis were found within breathing waveforms: a shorter duration of the breath cycle, a sharper abdominal peak vs. thoracic peak, a triangular shape to the abdominal peak, a rectilinear shape of the abdominal inhalation, and a sharp abdominal transition from inhalation to exhalation. The latter gave a particularly strong indication of neurosis. Strong signs of neurosis in men include dissimilarity in the recordings over the five days and a bell shape to the thoracic peak. In women, a lightly concave abdominal exhalation was indicative of neurosis. A somewhat less significant indicator for both men and women was an irregular-breathing pattern. Also, a lack of exhalation pauses also points to a neurotic condition.

Clausen comments on the variability of breathing patterns of neurotics: "Some authors…have claimed that neurotic patients have more deliberate controlled respiration which should be a means of suppressing emotions. The above findings may fit in with this assumption. The greater variability from day to day could indicate that neurotics do not breathe naturally and automatically, i.e., they have no characteristic basic type of response to which they might revert in a relaxed state. At the same time the deliberate control will guarantee a limited fluctuation around the level they are trying to maintain for that particular day" (Clausen, 1951, 51). Later in the paper, Clausen further summarizes: "Mental disturbance per se, whether neurosis or psychosis, changes the breathing pattern. More specifically, it could be that people with chronic emotional conflicts develop a cautious respiration, characterized by fast and shallow respiration of thoracic type as a means of suppressing the emotions" (Clausen, 1951, 60).

Respiration and Anxiety

While anxiety is the hallmark of a fully developed neurosis, many without a diagnosable neurotic condition suffer from it daily. Fritz Perls, the founder of Gestalt therapy, proposed a theory that explains anxiety in terms of inhibited breathing. One author (Fesmire, 1994) quotes Perls et al: "Anxiety is the experience of breathing difficulty during any blocked excitement" (Perls et al, 1951,128). In other words, we tend to block emotion by constricting our breathing. Perls felt that the societal taboo on showing fear causes us to block that emotion, resulting in an emotion closely related to fear-anxiety. When angry, we also try to keep control by holding our breath. Again Fesmire quotes Perls et al: "The anxiety occurs when we try to rein in the excitement within the limits of decorum" " (Perls et al, 1951, 129).

The link between anxiety and breathing was clearly shown in the following experiment. The respiratory efficiency of one group that suffered from anxiety and a control group free of any anxiety disorders were measured. Respiratory efficiency, which is "a measure of the respiratory effort required by the subject to extract 100ml of oxygen from the respired air" (Coppen and Mezey, 1960, 54), was significantly lower for the anxious group than for the control group. The anxious group was then injected with sodium amytal, a drug very effective in reducing anxiety. The respiratory efficiency of the anxious group became similar to that of the control group, suggesting a direct link between anxiety and respiration.

Some relief from anxiety may occur through the retraining of one's breathing patterns. One paper states that the retraining of one's breathing patterns can be used to treat sleeping difficulties, high/low muscle tension, and nervousness (Buckholtz, 1994, 147). Many psychologists suggest psychotherapy instead; the resulting emotional healing can then be demonstrated by changes in the breathing pattern.

Breathing Retraining

Imke Buckholtz (1994, 145) discusses the origin of breathing problems, stating that damaged breathing muscles are an obvious source. He further explains that "many breathing disorders have their origin in trauma, fear, shock, pain, grief, or other stressful events in both the early and adult years of life." And by suppressing one's breathing, a person can diminish the "pain, stress, or grief of a situation."

Buckholtz describes a method for retraining the breathing created by Elsa Gindler, who "experimented with movements to strengthen the deeper layers of the muscular system and improve the circulation of oxygen, movements that reduced tensions that had been preventing the breathing muscles from functioning properly" (Buckholtz, 1994, 141). He claims that breathing retraining often results in improved emotional stability (Buckholtz, 1994, 145). Thus the reverse of the axiom that emotions play a role in breathing patterns seems possible; changing the breathing results in an emotional change.

Buckholtz advises that Gindler's breathing retraining "does not directly address psychological problems" (Buckholtz, 1994, 146). He suggests that reduction of physical problems through breathing retraining may be what a person needs to make long-standing improvements in everyday life. However, he cautions: "But sometimes, especially with early childhood trauma, or if the client constantly finds himself falling back into the same restrictive patterns, the help of a psychotherapist is required. The patient needs to become aware of the primary experience that caused the physical block and 'undo' it by consciously going through it again" (Buckholtz, 1994, 146).

Changes in breathing that are consciously induced or conditioned by operant or classical means "can affect both our thoughts and our feelings" (Ley, 1994, 96). Ley also suggests "the implications of this premise are relevant to theory, diagnosis, and treatment of stress and anxiety-related disorders (e.g., panic disorder, phobias, test anxiety, occupational strain, and related psychosomatic disorders), and to basic and applied research in the psychophysiology of breathing" (Ley, 1994, 95).

Buckholtz's and Ley's ideas lend credence to the notion that changing the breath pattern aids in controlling emotional states. However, neither statement directly suggests creating this change through the use of monitoring devices such as the one described here.