Anticipating noxious stimulation rather than afferent nociceptive input may evoke pupil asymmetry

Unilateral nociceptive stimulation is associated with subtle signs of pupil asymmetry that may reflect lateralized activity in the locus coeruleus. To explore drivers of this pupil asymmetry, electrical stimuli, delivered alone or 200 ms before or after an acoustic startle stimulus, were administered to one ankle under four experimental conditions: with or without a 1.6 s anticipatory period, or while the forearm ipsilateral or contralateral to the electrical stimulus was heated tonically to induce moderate pain (15 healthy participants in each condition). Pupil diameter was measured at the start of each trial, at stimulus delivery, and each second for 5 s after stimulus delivery. At the start of the first trial, the pupil ipsilateral to the side on which electric shocks were later delivered was larger than the contralateral pupil. Both pupils dilated robustly during the anticipatory period and dilated further during single-and dual-stimulus trials. However, pupil asymmetry persisted throughout the experiment. Tonically-applied forearm heat-pain modulated the pupillary response to phasic electrical stimuli, with a slight trend for dilatation to be greater contralateral to the forearm being heated. Together, these findings suggest that focusing anxiously on the expected site of noxious stimulation was associated with dilatation of the ipsilateral pupil whereas phasic nociceptive stimuli and psychological arousal triggered bilateral pupillary dilatation. It was concluded that preparatory cognitive activity rather than phasic afferent nociceptive input is associated with pupillary signs of lateralized activity in the locus coeruleus.


Introduction
Neural circuits that traverse the locus coeruleus modulate pain and other sensations at various sites including the spinal cord, thalamus and cerebral cortex (Llorca-Torralba et al., 2016;Suárez-Pereira et al., 2022;Szabadi, 2012).By-and-large this modulation is inhibitory (McBurney-Lin et al., 2019) and, at least within the spinal cord, is stronger ipsilateral than contralateral to the side of nociceptive input (Tsuruoka et al., 2003).The inhibitory influence extends to spinal segments distant from the site of pain, consistent with the presence of a broadly distributed but lateralized modulatory system (Tsuruoka et al., 2004).
Afferent nociceptive input to the locus coeruleus, and top-down inputs stemming from threat-processing centres such as the prefrontal cortex and amygdala, trigger pupillary dilatation in step with locus coeruleus activity (Murphy et al., 2014;Szabadi, 2012).In lightly anaesthetised rodents, the ipsilateral pupil dilates more than the contralateral pupil during unilateral electrical micro-stimulation of the locus coeruleus (Liu et al., 2017).In conscious humans, bilateral pupillary effects of arousal appear to predominate during painful stimulation (Drummond, 2022;Drummond and Clark, 2023).Nevertheless, subtle signs of an ipsilateral component have been identified.For example, during painful cold pressor tests, the ipsilateral pupil redilates faster than the contralateral pupil after a momentary flash of light (Drummond, 2022).
The main aim of this study was to further explore potential influences on pupil asymmetry.Micro-stimulation of the locus coeruleus triggers a steep but transient increase in pupil size whereas more prolonged activation induces an extended gradual dilation that, at low frequencies, facilitates dilatation to phasic stimuli (Liu et al., 2017).Hence, in the present study, it was hypothesized that momentary electrical stimulation of the ankle (thereby activating the locus coeruleus) would trigger pupil asymmetry and that heating the forearm to induce additional pain would facilitate this response.To investigate ancillary effects of psychological arousal on the pupillary response, some stimuli were preceded by an anticipatory period; in addition, electrical stimuli were either presented alone or delivered shortly before or after an acoustic startle stimulus.As the pupils dilate during psychological arousal, it was hypothesized that anticipating the stimulus and pairing the electrical stimulus with acoustic startle would mask effects of unilateral nociceptive stimulation.
Whether these experimental manipulations influenced the perceived intensity or unpleasantness of electrical or acoustic startle stimuli was also examined.Specifically, it was hypothesized that heterotopic noxious counter-stimulation induced by heating the forearm would inhibit electrically-evoked pain (Bannister and Dickenson, 2016;Ladouceur et al., 2012) and that inhibition would be stronger for ipsilateral than contralateral conditioning stimuli (Tsuruoka et al., 2004).Effects of emotion on the perception of pain are complex (Rhudy and Meagher, 2000); hence, no specific predictions were made about the influence of anticipation on the perceived intensity or unpleasantness of electrical stimuli.However, as the first stimulus inhibits perception of a second stimulus presented 200 ms later (English and Drummond, 2021), it was hypothesized that leading with an acoustic startle stimulus would inhibit electrically-evoked pain when these stimuli were presented close together.

Participants
The sample consisted of 29 healthy males and 31 healthy females aged between 18 and 51 years (mean ± standard deviation 25 ± 8 years) who were recruited via a study enrolment website and through personal contacts.Pregnant or breastfeeding women, people who were being treated medically for a physical or psychiatric disorder, who suffered from hearing loss or who had epilepsy or a pacemaker, were not eligible to participate.Participants provided their informed written consent for the procedures, which were approved by the Murdoch University Human Research Ethics Committee.In recognition of their participation, participants received a coffee voucher or were credited for experimental participation in their undergraduate psychology course.

Procedures
The procedures were carried out in a temperature-controlled laboratory maintained at 22 ± 1 • C. The experimental sequence is summarised in Fig. 1.

Experimental conditions (Fig. 1)
Participants were pre-allocated via a random number generator to one of four experimental conditions with 15 participants in each condition: stimulus delivery after an anticipatory period (the Control condition); stimulus delivery without an anticipatory period (the Immediate Stimulus Delivery condition); heating the ipsilateral forearm before and during an anticipatory period and after stimulus delivery (Ipsilateral Heating); and an identical sequence while heating the contralateral forearm (Contralateral Heating).During the anticipatory period in the Control, Ipsilateral Heating and Contralateral Heating conditions, eight sequential camera shutter clicks at 0.2 s intervals heralded the imminent delivery of noxious stimulation (see Section 2.2.4).

Stimuli (Fig. 1)
Five trials were presented in each of the following stimulus categories: electrical stimulus alone; acoustic stimulus to the ear ipsilateral or contralateral to the electrical stimulus; electrical stimulus followed 200 ms later by an acoustic stimulus to the ipsilateral or contralateral ear; acoustic stimulus to the ipsilateral or contralateral ear followed 200 ms later by an electrical stimulus.These single and dual stimuli were presented in random order at 1-2-min intervals.
The electrical stimulus consisted of a train of ten 0.5 millisecond (ms) monopolar square-wave pulses delivered at 5 ms intervals via a custombuilt concentric electrode (i.e., the stimulus train lasted 50 ms).The electrode (a centrally placed wire cathode surrounded by an annular stainless-steel anode with an inner diameter of 10 mm and an outer diameter of 20 mm) was coated with conductive gel and attached with an adhesive washer to the ankle between the left or right lateral malleolus and the Achilles tendon.Pulses were generated by a DS7A Digitimer constant current stimulator (Welwyn Garden City, UK).Starting at 1.0 mA, current strength was adjusted up or down in 0.5 mA steps and then 0.1 mA steps until given a pain rating of 5 on a 0 to 10 scale of pain intensity.This stimulus intensity was used thereafter.The acoustic startle stimulus was a 50-ms burst of atonal noise (110 dB SPL, instantaneous rise time), directed to the left or right ear via Sennheiser HD201 headphones (Chatswood, New South Wales, Australia).

Forearm heating conditions
Heat was delivered via a purpose-built thermode (2 cm diameter contact area) applied to the volar aspect of the left or right forearm.Before the first trial, the temperature of the thermode was adjusted individually to evoke a pain rating of 5 on a 0-10 numerical pain rating scale (i.e., a moderately painful sensation).During this process, the temperature of the thermode increased in 1 • C steps from 41 • C at 20 s intervals until moderate pain was reported.This temperature was used thereafter.The heated thermode was applied to the volar aspect of the forearm ipsilateral or contralateral to the concentric electrode for 20 s before the trial started and was removed ~5 s after the electrical and/or acoustic stimulus was delivered.To prevent sensitisation of the forearm, the thermode was placed on a slightly different site during each trial.

Pupil assessment
The participant sat with their forehead and chin supported in front of an opaque screen illuminated at 100 lx (measured with a Litemate III photometer, Kollmorgan Corporation, Burbank CA at the position of the participant's eyes).A Nikon D7000 camera (Nikon Corporation, Tokyo, Japan), modified to extend into the infrared range to enhance the contrast between the pupil and iris, was positioned in front of the screen.The pupils were illuminated by infrared light sources out of the participant's direct line of sight.In each trial, the pupils were photographed 5 times/s for 7 s.Stimuli were delivered at the onset of the photographic Fig. 1.Experimental conditions and stimuli.Participants were assigned randomly to an Immediate Stimulus Delivery condition or to one of three anticipatory conditions (control; heating the forearm ipsilateral to electric shock delivery before and during stimulus presentation; or heating the contralateral forearm before and during stimulus presentation) (15 participants in each condition).Single and dual electrical and acoustic startle stimuli were administered in random order.In each trial, photographs of the pupils were taken 5 times/s for 7 s.In the anticipatory conditions, the first stimulus (S1) was presented when the 8th photograph was taken, and the second stimulus (S2) was presented 0.2 s later.In the Immediate Stimulus Delivery condition, the first stimulus (S1) was presented when the 1st photograph was taken, and the second stimulus (S2) was presented 0.2 s later.

P.D. Drummond
sequence in the Immediate Stimulus Delivery condition but were administered 1.6 s after the first photograph (i.e., after the anticipatory period) in the other three experimental conditions (Fig. 1).During the anticipatory period, repetitive camera shutter clicks signalled the imminent arrival of an electrical and/or acoustic stimulus.

Ratings
After each electrical stimulus, pain intensity and unpleasantness were rated verbally between 0 (indicating none) and 10 (indicating extreme).The loudness of acoustic stimuli and acoustic discomfort were rated on similar scales.For dual stimuli, pain ratings preceded acoustic ratings.

Data reduction and statistical approach 2.3.1. Pupil diameter
Pupil diameter was measured from the first photograph in each trial, at stimulus delivery, and each second for 5 s after stimulus delivery using a Matlab script or the Adobe Photoshop ruler tool (San Jose, CA) if the script failed to identify the pupils.If pupil diameter could not be measured (e.g., because of blinking), the next image in the sequence was used (i.e., 0.2 s later; 0.2 % of measures).In each image, the pupils were calibrated against a 6.2 mm diameter sticker attached to the forehead or bridge of the nose.These measures were averaged across the five trials in each stimulus category.

Ratings
Each type of rating was averaged across the five trials in each stimulus category.Pain intensity and unpleasantness to electric shocks delivered alone were compared among the four experimental conditions in analyses of variance (ANOVA).Ratings of loudness and auditory discomfort to single acoustic stimuli were investigated in Experimental Condition x Acoustic Stimulus Side (ipsilateral versus contralateral to the side on which electric shocks were delivered) repeated measures ANOVAs.To examine dual-stimulus effects, differences in ratings between single-and dual-stimulus trials were calculated.These difference scores were investigated in Experimental Condition × Acoustic Stimulus Side × Stimulus Order (shock first or acoustic stimulus first) repeated measures ANOVAs.

Analyses of pupil diameter
Pupil diameter at the onset of each trial was investigated in Experimental Condition x Pupil Laterality (ipsilateral versus contralateral to the side on which electric shocks were delivered) × Trials (the 35 consecutive trials).In the Control, Ipsilateral Heating and Contralateral Heating conditions, changes in pupil diameter during the 1.6 s anticipatory period were investigated in Experimental Condition × Time (baseline versus stimulus delivery) × Pupil Laterality (ipsilateral versus contralateral to the side on which electric shocks were delivered) × Trials (the 35 consecutive trials).In addition, changes in pupil diameter after stimulus delivery were investigated in Experimental Condition × Time (each second for 5 s after stimulus delivery) × Pupil Laterality repeated measures ANOVA.Where appropriate, the ANOVA also included factors of Stimulus Order (shock first or acoustic stimulus first) and Laterality of the Acoustic Stimulus.Changes in pupil diameter after stimulus delivery were investigated in similar analyses in the Immediate Stimulus Delivery condition.

Planned contrasts
Significant main effects and interactions that included Experimental Condition were investigated using planned contrasts: (a) between the two heating conditions and the Control condition (to determine whether forearm heating influenced ratings or pupillary responses); (b) between the Ipsilateral and Contralateral Heating conditions (to determine whether heating laterality influenced ratings or pupillary responses); and (c) between the Control and Immediate Stimulus Delivery conditions (to determine whether anticipating stimulus delivery influenced ratings or pupillary responses).
SPSS version 28 was used to analyse the data.In all analyses, the multivariate solution was used to avoid violation of the sphericity assumption, and the criterion of statistical significance was p < .05. Results are presented as the mean ± standard error in the text and figures.Effect sizes are reported as (partial) eta squared (η 2 ), which was equivalent to Pillai's trace in the repeated measures ANOVAs reported below.

Results
Significant influences on ratings and pupil diameter are presented below.Complete tables of inferential statistics are presented in a supplementary file.
In single-stimulus trials, pain and auditory ratings were similar in each experimental condition (Table 1).Notably, neither heating the ipsilateral nor the contralateral forearm inhibited electrically evoked pain.
In dual-stimulus trials, neither stimulus order nor the laterality of the acoustic startle stimulus influenced pain intensity ratings in the groupas-a-whole (Fig. 2A).However, the acoustic pre-pulse inhibited electrically-evoked pain in the Control condition (p = .012),thus replicating previous findings (English and Drummond, 2021).Likewise, under some but not all experimental conditions, the acoustic pre-pulse inhibited pain unpleasantness [Experimental Condition × Stimulus Order interaction: partial η 2 = 0.133; F(3, 56) = 2.85, p = .045](Fig. 2B).Investigation of this effect with planned contrasts indicated that inhibition of pain unpleasantness by the acoustic pre-pulse was greater in the Control than forearm heating conditions (p = .043)(Fig. 2B).
For auditory ratings, the electrical pre-pulse inhibited loudness and auditory discomfort evoked by the acoustic startle stimulus [main effect for Stimulus Order: for loudness, partial η 2 = 0.191; F(1, 56) = 13.2, p < .001;for auditory discomfort, partial η 2 = 0.268; F(1, 56) = 20.5, p < Note: Ratings were similar in each experimental condition and were similar for acoustic stimuli delivered ipsilateral or contralateral to the side on which electric shocks were delivered.
In the three anticipatory conditions, the pupils dilated symmetrically during the anticipatory period before stimulus delivery [from 4.41 ± 0.10 mm at baseline to 4.64 ± 0.10 mm 1.6 s later, main effect for Time, partial η 2 = 0.681; F(1, 42) = 89.9,p < .001](Fig. 3B).This dilatation was similar in each anticipatory condition and persisted across the 35 trials.At the end of the anticipatory period of the first trial, immediately before delivery of the first stimulus, the ipsilateral pupil was significantly larger than the contralateral pupil [5.01 ± 0.71 mm versus 4.86 ± 0.76 mm, t(44) = 3.19, p = .003].

After stimulus delivery: Immediate Stimulus Delivery condition
Both pupils dilated robustly after single and dual electrical and acoustic stimuli, peaking 1-2 s after stimulus delivery (Fig. 4A-G).Pupillary dilatation was symmetrical during single-stimulus trials (Fig. 4A-C).In dual-stimulus trials, stimulus order and laterality of the acoustic startle stimulus moderated the extent of pupillary dilatation [interaction between Pupil Laterality, Time, Laterality of the Acoustic Stimulus and Stimulus Order: partial η 2 = 0.570; F(4, 11) = 3.65, p = .040](Fig. 4D-G).Exploration with planned contrasts failed to clarify the source of this interaction.However, response asymmetry appeared to be greater when the electric shock was followed by an ipsilateral acoustic startle stimulus than in the other stimulus combinations (Fig. 4D-G); relative to other stimulus combinations, dilatation of the contralateral pupil was slightly greater than dilatation of the ipsilateral pupil (Fig. 4D).

After stimulus delivery: Control, Ipsilateral Heating and Contralateral Heating conditions
After strong dilatation during the anticipatory period, further minor increases in pupil diameter peaked ~2 s after delivery of single and dual electrical and acoustic stimuli (Figs. 5 to 7).Pupillary responses to electrical stimulation of the ankle differed among the three experimental conditions [Experimental Condition × Time × Pupil Laterality interaction, partial η 2 = 0.364; F(8, 80) = 2.22, p = .034](Fig. 5).
Exploration with planned contrasts failed to clarify the source of this interaction.However, in terms of pupil asymmetry, forearm heating appeared to modulate pupillary dilatation to the electrical stimulus with effects peaking contralateral to the arm being heated (Fig. 5A-C).Pupillary responses to single acoustic stimuli and to dual electrical and acoustic stimuli were similar in each experimental condition (Figs. 6 and  7).

Discussion
The main aim of this study was to explore drivers of pupil asymmetry to painful stimulation.The pupil ipsilateral to the side on which electric shocks were delivered was larger than the contralateral pupil throughout the experiment.This asymmetry was unaffected by anticipating noxious stimulation and was only marginally affected by delivery of single or dual electrical and acoustic stimuli and by pain induced by heating the forearm ipsilateral or contralateral to the side on which electric shocks were delivered.

Potential mechanism of pupil asymmetry
A trend for the ipsilateral pupil to be larger than the contralateral pupil was detected before the first stimulus was delivered and persisted thereafter.In our previous work, a similar trend was identified not only during a painful cold pressor test but also during prior immersion of the hand in lukewarm water (Drummond and Clark, 2023).The effect weakened after repeated immersions, arguing against straightforward mediation by a somato-autonomic reflex that relayed directly through the locus coeruleus.Given this, it seems plausible that influences other than lateralized nociceptive afferent input to the locus coeruleus contribute to pupil asymmetry.Candidates could include focal activity in threat-processing centres that communicate with the locus coeruleus, such as the amygdala and prefrontal cortexthat is, efferent rather than afferent input to the locus coeruleus might initiate lateralized activity.
Clusters of neurons within the locus coeruleus appear to function as modules that interact with specialized cortical, subcortical and spinal centres to shape a variety of behaviours (Chandler et al., 2019).In one well-documented example of this property, stimulation of neurons in the locus coeruleus that projected to the prefrontal cortex provoked signs of anxiety in rodents whereas stimulation of spinally-projecting neurons induced analgesia (Hirschberg et al., 2017).This modular organisation In this and subsequent figures, results are presented for pupils ipsilateral (filled circles) and contralateral (unfilled circles) to the concentric electrode attached to the ankle (the source of electric shocks).The photographic images illustrate pupil asymmetry at baseline.The ipsilateral pupil was larger than the contralateral pupil for most of the experiment.(B) In the three anticipatory conditions, both pupils dilated in anticipation of the electrical and acoustic stimuli.The photographic images, taken 1.6 s after the first photograph in the sequence, illustrate dilatation of both pupils during the anticipatory period.may also extend to nociceptive controlsneurons in the ventral locus coeruleus that project to the spinal dorsal horn form part of the endogenous antinociceptive control system whereas more dorsally-sited neurons facilitate nociception (Hickey et al., 2014).Whether an interaction between these modules influences lateralized autonomic activity is uncertain, but it would be interesting in future studies to systematically manipulate attentional focus and anxiety during the delivery of nociceptive stimuli to explore their individual and combined effects on pupil asymmetry.For example, focusing anxiously on the expected site of noxious stimulationas might be envisaged at the start of this experimentcould explain why the ipsilateral pupil was larger than the contralateral pupil before delivery of the first stimulus.Intriguingly, focusing attention on one side of the face intensifies blushing on that side (Drummond and Mirco, 2004), consistent with a link between lateralized cognitive and autonomic activity.
In studies on rodents, Liu et al. (2017) reported that dilatation of the ipsilateral pupil was significantly greater than dilatation of the contralateral pupil during unilateral tonic 2 and 5 Hz stimulation of the locus coeruleus; in addition, an inverted U-shaped association between background stimulation frequency and pupillary dilatation to additional phasic stimulation was identified, peaking at 2 Hz.In the present study, tonically applied heat-pain induced minor changes in the pupillary response to phasic electrical stimuli contralateral to the arm being heated.Likewise, in the Immediate Stimulus Delivery condition, dilatation of the contralateral pupil was slightly greater than dilatation of the ipsilateral pupil when the electric shock was followed immediately by an ipsilateral acoustic startle stimulus.The mechanism of these lateralized effects is uncertain, but it might be relevant that axons from second-order neurons in the spinal dorsal horn decussate before transmitting nociceptive and thermal information to the locus coeruleus (Craig, 1995;Westlund and Craig, 1996).Hence, it might be expected that afferent nociceptive input would preferentially stimulate the contralateral locus coeruleus (Voisin et al., 2005), and that spatial summation of inputs would amplify this effect.However, this is at odds with the observation that pinching a paw in two anaesthetised rodents triggered greater dilatation of the ipsilateral than contralateral pupil  P.D. Drummond (Liu et al., 2017).Whether this was due to afferent input or to top-down influences on locus coeruleus activity was not explored.

Perception of nociceptive and acoustic startle stimuli
An additional aim of this research was to determine whether anticipating noxious stimulation and/or heterotopic noxious counterstimulation influenced the perception of electrical stimuli.In the conditioned pain modulation paradigm, stronger pain inhibits weaker pain, presumably to prioritise responding to the most likely source of harm.This effect varies between individuals owing, in part, to differences in mood and/or salience of the painful stimulus (Nahman- Averbuch et al., 2016;Traxler et al., 2019).In the present study, neither anticipating the stimulus nor heating the forearm to induce moderate pain influenced ratings of the intensity or unpleasantness of electrical stimuli.This might have been due to a floor effect because, by the end of the experiment, electrical stimuli were rated as only mildly painful and unpleasant.
In humans, cold pressor tests, heating capsaicin-sensitized skin and high frequency electrical stimulation not only induce signs of central sensitization but also trigger analgesia in the ipsilateral forehead (Knudsen and Drummond, 2009;Knudsen and Drummond, 2011;Vo and Drummond, 2013).Presumably, such effects are mediated by lateralized coeruleo-spinal inhibitory controls (Tsuruoka et al., 2004;Tsuruoka et al., 2003); thus, inhibitory effects of noxious counterstimulation were expected to be stronger ipsilateral than contralateral to electrical stimulation.However, this was not confirmed, possibly due to the floor effects noted above or to lack of central sensitization in the current paradigm.It is worth noting that pupil diameter in the forearm heating conditions was similar to pupil diameter in the control condition that is, there was no sign that heat-pain increased tonic activity in the locus coeruleus.Thus, a stronger conditioning stimulus might be required to evoke inhibitory coeruleo-spinal effects on pain.A further consideration is that, in rodents, bilateral lesions of the locus coeruleus are required to block inhibitory coeruleo-spinal influences on hyperalgesia associated with inflammation (Maeda et al., 2009).Hence, topdown rather than lateralized bottom-up processes might drive coeruleo-spinal pain inhibitory effects.
In our previous work, an acoustic startle stimulus delivered after an anticipatory period of 1.6 s inhibited pain to an electrical stimulus presented 200 ms later; conversely, when administered in reverse order, the electrical stimulus inhibited loudness and discomfort to the acoustic startle stimulus (English and Drummond, 2021).In the present study, the acoustic pre-pulse inhibited shock-evoked pain in the control condition but not when the anticipatory period was omitted or when heat was applied to the forearm.The inhibitory influence of pre-pulses may assist information processing of the pre-pulse by reducing interference from extraneous stimuli (Swerdlow et al., 2005).However, in the present study, focusing attention on the pre-pulse might have been difficult when stimuli were delivered without warning or when the forearm was heated.The failure of electrical pre-pulses to inhibit loudness or auditory discomfort to acoustic startle stimuli in the Immediate Stimulus Delivery condition supports this interpretation.Curiously, the inhibitory influence of the electrical pre-pulse on auditory discomfort was greater  when the contralateral than ipsilateral forearm was heated.Why this was so is unclear, but perhaps heat applied to the contralateral arm was less distracting, as it was well away from the focus of attention, than heat applied to the ipsilateral forearm.Likewise, in the control condition, a sudden switch in attention to the contralateral ear might explain why the pre-pulse failed to inhibit loudness and auditory discomfort to contralateral acoustic startle stimuli.

Limitations
A limitation of this study was the use of only a single intensity of electrical stimulation, adjusted individually at the start of the experiment to evoke moderate pain (a rating of 5 on a 0-10 scale of pain intensity).However, ratings declined in later trials.Given this, the association between pain intensity and pupil asymmetry requires further study.
As this was the first study of its kind, a power analysis to estimate an appropriate sample size was not possible.Thus, weak effects might have been missed.
Furthermore, pupil diameter was not assessed before preliminary tests to identify the current strength that evoked moderate pain.To investigate whether pupil asymmetry developed during this procedure, changes in pupil diameter were assessed in 11 volunteers who did not participate further in the study.Pupil asymmetry did not develop (data not reported), suggesting the involvement of some other mechanism.However, this needs to be explored further.

Conclusions
Overall, the experimental manipulations appeared to influence the perception of nociceptive and acoustic stimuli more strongly than pupil reactivityratings of stimulus intensity and unpleasantness to dual stimuli varied among the four experimental conditions whereas pupillary responses were similar on both sides.In terms of pupil reactivity, it seems likely that the ipsilateral pupil dilated when participants focused anxiously on the expected site of noxious stimulation.Thereafter, the nociceptive stimulus triggered minor dilatation of the contralateral pupil whereas both pupils dilated strongly when participants anticipated noxious stimulation.Thus, under the conditions of this experiment, cortical influences appeared to overshadow any lateralized effect of afferent nociceptive input on pupil diameter.

Declaration of competing interest
The author does not report a conflict of interest.This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

.
001,] but, once again, this differed among the four experimental conditions [for loudness, Experimental Condition × Laterality of the Acoustic Stimulus × Stimulus Order interaction: partial η 2 = 0.141; F(3, 56) = 3.06, p = .036;for auditory discomfort, Experimental Condition × Stimulus Order interaction: partial η 2 = 0.171; F(3, 56) = 3.84, p = .014](Fig. 2C and D).Investigation of these interactions with planned contrasts indicated that the inhibitory influence of electrical pre-pulses on auditory discomfort was greater in the Contralateral than Ipsilateral Heating condition [Experimental Condition (ipsilateral versus contralateral heating) × Stimulus Order interaction, p = .028].In addition, the electrical pre-pulse inhibited loudness and auditory discomfort to ipsilateral startle stimuli in the Control condition, but failed to inhibit loudness or auditory discomfort to ipsilateral or contralateral startle stimuli in the Immediate Stimulus Delivery condition [for loudness: Experimental Condition (control versus combined heating) × Laterality of the Acoustic Stimulus × Stimulus Order interaction, p = .039;and Experimental Condition (control versus immediate stimulus delivery) × Laterality of the Acoustic Stimulus × Stimulus Order interaction p = .015;for auditory discomfort: Experimental Condition (control versus combined heating) × Laterality of the Acoustic Stimulus × Stimulus Order interaction, p = .021](Fig. 2C and D).

Fig. 2 .
Fig. 2. Differences ± standard error between dual-and single-stimulus trials of ratings for (A) the intensity and (B) unpleasantness of the painful electrical stimulus, and for (C) loudness and (D) auditory discomfort evoked by the acoustic startle stimulus.

Fig. 3 .
Fig. 3. (A) Pupil diameter ± standard error at baseline (the first photograph in each photographic sequence).In this and subsequent figures, results are presented for pupils ipsilateral (filled circles) and contralateral (unfilled circles) to the concentric electrode attached to the ankle (the source of electric shocks).The photographic images illustrate pupil asymmetry at baseline.The ipsilateral pupil was larger than the contralateral pupil for most of the experiment.(B) In the three anticipatory conditions, both pupils dilated in anticipation of the electrical and acoustic stimuli.The photographic images, taken 1.6 s after the first photograph in the sequence, illustrate dilatation of both pupils during the anticipatory period.

Fig. 4 .
Fig. 4. Change ± standard error in pupil diameter in the Immediate Stimulus Delivery condition to (A) single electrical stimuli; (B, C) single acoustic stimuli; and (D-G) dual electrical and acoustic stimuli.

Fig. 5 .
Fig. 5. Change ± standard error in pupil diameter in the three anticipatory conditions 1-5 s after delivery of electrical stimuli.

Fig. 6 .
Fig. 6.Change ± standard error in pupil diameter in the three anticipatory conditions 1-5 s after delivery of acoustic startle stimuli.P.D. Drummond

Table 1
Ratings during single-stimulus trials.