Abstract
Although people have made clay pots for millennia, little behavioral research has explored how they do so. We were specifically interested in potters’ use of auditory, haptic, and visual feedback. We asked what would happen if one or two of these sources of feedback were removed and potters tried to create pots of a given height, stopping when they thought they had reached that height. We asked students in a pottery class to build simple clay vessels either when they had full sensory feedback (in the control condition for all participants) or when they had reduced input from one modality (in Experiment 1) or two modalities (in Experiment 2). Participants were asked to stop building the vessels when they thought the vessels were 5 in. high. We found that participants produced shorter vessels when one or more forms of sensory feedback was reduced. The degree of shortening did not depend on the type or number of reduced sensory channels. The results are consistent with a control hypothesis where potters must have learned how to use sensory feedback from the modalities to help them control their ceramic creations. The results help highlight the importance of the intimate connections between perception and action.
Introduction
Interaction with the world relies on multisensory input. Understanding how those multiple sensory inputs combine is one of the core aims of research in perception. Several paradigms, such as sensory suppression (Juravle & Spence, 2011; Ziat, Hayward, Chapman, Ernst, & Lenay, 2010) and multisensory incongruity (Spence, Pavani, & Driver, 2004; Ziat, Balcer, Shirtz, & Rolison, 2016), have helped shed the light on the way the senses are integrated. Relatedly, some product design researchers have explored the impact of removing or reducing sensory input on product choice (Schifferstein, Otten, Thoolen, & Hekkert, 2010). We wondered, therefore, how greater reliance on one or more modalities, if indeed that occurred following the attenuation of other modalities, would affect potting performance. In this study, we exploited research of the third kind (modality reduction) to investigate a kind of skill that relies on multisensory integration in ways that have not been previously explored. The kind of skill we investigated is one that humans have engaged in for thousands of years—the creation of clay pots.
Through the thousands of years that people have made clay pots, they have had access to visual, tactile, auditory, and olfactory cues as they formed the pots. 1 We wondered how the visual, tactile, and auditory cues are used. We refrained from exploring smell in this study, however. We were attracted to the problem of the way sight, touch, and hearing are used in potting because potting, despite its centrality as a form of cultural expression for so long, has been subjected to very little scientific behavioral research, though a bit of work has been done on virtual pottery (Han & Han, 2014) and visuo-tactile matching tasks, where potters’ tactual abilities were found to be slightly better than those of nonpotters (Power & Graham, 1976). We found potting to be interesting as a test case for exploring how multimodal perception guides skilled action. The possibility of exploring multimodal perception in the skill of potting is also interesting because it affords an ecologically valid way of investigating the inextricable coupling of perception and action (Gibson, 1979).
Aiming to study this important coupling, we introduced two distinct hypotheses about how the senses combine in potting. The hypotheses we developed relate to broader issues in the understanding of perception–action relations. This study adds to work done by others on the way that multisensory input guides skilled performance (e.g., Esteves & Spence, 2014).
The specific impetus for our study of pottery arose, literally, from an accident a potter student who had burned her hand (outside of class) told the first author. While recovering from the burn, the student wore gloves while potting. The student reported that she relied more on the sound of the potter wheel than she had before as a way of compensating for the reduced cutaneous feedback. Others have studied the effects of wearing gloves on performance involving tactile sensitivity (e.g., Heller & Mitchell, 1985).
To approach the hypotheses we developed, we want to say a bit more about pottery itself. When potters make ceramic pots at a potter’s wheel (the environment of interest here), they get rich multimodal information over the course of building their vessels. They see the clay in and around their hands, they hear the whir of the wheel, they feel the clay on their fingers, and they smell the clay. Throwing a pot—the term for making a clay vessel on a wheel—relies on all of these sources of information in a process of active interaction between material and maker. Throwing a pot has five main steps:
Wedge the clay: Once a lump of clay is placed on the wheel, it needs to be firmly secured to start centering process. Center the clay: Maintain the clay in the center while beginning to shape it. Establish a base: Once centering has been achieved, press the clay (typically with the two thumbs) toward the center of the cylinder to form the aperture of the vessel. Bring up the walls: Bring the vessel up to a height that depends on the pressure applied by one hand pressing in and the other hand pressing out to support the interior. Stop: Once the pot is as tall as the potter wants, terminate the building process. How tall the pot gets depends on the potter’s aesthetic or utilitarian goals, and on how controllable the building process is perceived to be. If the pot has no predetermined height, then stopping occurs when control of the pot seems too difficult—that is, when the height is greater than the height that can be managed.
What roles do seeing, hearing, and feeling play in the just-described process? Because this study was the first on this topic, as far as we know, we did not attempt to characterize the selective effects that the modalities might have on each of the five steps. Rather, we sought to determine the height at which potters stopped when they had full sensory information as opposed to when they had reduced sensory information about each of the modalities singly (in Experiment 1) or in pairs (in Experiment 2). In all of the conditions of both experiments, we asked our potters to stop when they thought their pots were 5 in. high (12.7 cm). That height was familiar to them; it was the height they were asked to produce during the final assignment in the class we studied. (Because the students were trained and worked with inches, we report those values but also provide the equivalent units in centimeters.)
We considered two main hypotheses. According to one, the reference condition for producing the height of the pot would be the sum of the inputs received, suitably transformed to some amodal or commensurate unit of measure, in which case potters would make taller pots when a source of feedback was reduced (the compensation hypothesis). For example, if potters could not feel the clay as much as in normal conditions or could not hear the wheel as much as in normal conditions, they would make the pots taller to increase the greater seen height. Such compensation has been observed in other domains. For example, in handwriting, removal of visual feedback leads to enlargement of written output (Potgieser, Roosma, Beudel, & de Jong, 2015). If the compensation depended on the type and number of withdrawn modalities, the potters would make the pots taller when two modalities were withdrawn than when only one modality was withdrawn. Furthermore, if the loss of information about the two modalities had additive effects, the combined added height would be equal to the excess height associated with one modality’s removal plus the excess height associated with the other modality’s removal.
According to the alternative view, the reference condition for judging the height of pot would be the degree of control that could be achieved as the pot grew, in which case the potters would make shorter pots when a modality was attenuated, reflecting perceived loss of control (the control hypothesis). The rationale for this prediction was that the shorter the pot, the smaller the centrifugal force associated with it, so the easier its control. By this hypothesis, the degree of shortening would index the degree to which the potters felt out of control when a given modality was suppressed. If the perceived loss of control grew additively with the loss or suppression of multiple modalities, the shortening associated with the suppression of those modalities would be equal to the sum of the shortenings associated with suppression of each modality on its own.
Experiment 1
In the first experiment, we reduced input of one of the three modalities of interest (audition, haptics, and vision), leaving the other two modalities intact.
Method
Participants
Twenty-four students at Northern Michigan University (16 females and 8 males, mean age = 24 years, standard deviation = 7.97 years) took part in this study. All participants were an introductory ceramics class. Their number was limited by the number of pottery wheels in the classroom where the study was conducted. Because we did not have any control on the level of expertise, as it depended on the students who happened to enroll each semester, we waited 2 months after the semester start to run the experiment so beginner potters could work independently while operating the wheel. This time was based on the third author’s years of experience teaching pottery. Participants gave informed consent before participating, and the procedure was approved by the university Institutional Review Board (HS13-540).
Apparatus and task
All participants used a Brent model C pottery wheel (Figure 1). The 14-in. (35.56 cm) aluminum wheel head was fitted with a 12-in. (30.48 cm) throwing bat. This model has a permanent magnet ¾ horsepower, 7 amp direct current motor, which offers a smooth speed range of 0 to 240 rpm. In the experiment, participants were given a pugged cylinder of stoneware clay (1.5 pounds or 680 g), a gallon bucket of water, and a small sponge. The task was to shape a 5 in. (12.7 cm) tall cylindrical vessel. This is the first type of form that beginner potters try to make. In the class, outside the context of this experiment, the students were required at the end of the semester to perform a similar task, except that for their exam, their sensory inputs were not reduced, and they had a ruler to measure their pot during shaping. They did not have a ruler in our experiment. Both in the exam and in our experiment, the students were given no requirement about the diameter or thickness of the pot. Those aspects are usually requested at upper levels of ceramics instruction.
A potter shaping a vessel on the wheel (left panel) and the pot being measured after its completion (right panel). The ruler was not present while the pot was being constructed.
After the experiment, participants were asked about their years of practice. Seventeen participants were beyond beginner status but not yet expert (less than 1 year experience). Five participants had an intermediate level of experience (less than 5 years). The remaining two participants were considered experts. The assignment of participants to one group or another was random, and no specific training was given for this experiment. Pictures of each pot were taken for final validation of the height and diameter (see Figure 1).
Procedure
The participants were randomly divided into three groups that differed with respect to the modalities that were fully available to them. The assignment of participants to groups was random as well with respect to the students’ degree of prior experience, which was not ascertained until after the experiment was over based on debriefing. A posteriori evaluation of the frequency of students with different degrees of expertise being assigned to the groups confirmed equivalent assignment.
In terms of what the groups were, one group had full auditory and haptic input but reduced visual input. These subjects wore a blindfold. Another group had full visual and haptic input but reduced auditory input. These subjects wore earplugs and headphones with a noise reduction rating of 62 dB. A third group had full auditory and visual input but reduced haptic input. These participants wore Nitrile gloves (15 mil or 0.381 mm thick) that significantly reduced the cutaneous contact with the clay. (Nitrile gloves are nonallergenic gloves similar to latex.)
Besides performing with reduced input, each participant also performed with full input. The order of the control condition (full input) and experimental condition (reduced input) was randomized over participants. After each pot was completed, the experimenter measured the pot out of sight of the participant and did not tell the participant how tall the pot was.
Results
The dependent variables were the times to complete the throws of the cylinders on the wheel, the resulting heights, and the cylinders’ final diameters. The test of homogeneity of variance was not violated for any of the three variables (p > .05) for any condition. For the test condition, the one-way analysis of variance (ANOVA) showed no significant differences between modalities for time, F(2, 23) = .43, p = .67; height, F(2, 23) = .1, p = .99; or diameter, F(2, 23) = .53, p = .6. Similarly, no significant difference was found for the control condition for time, F(2, 23) = 2.22, p = .13; height, F(2, 23) = .23, p = .79; or diameter, F(2, 23) = .1, p = .92. The only significant result was for height between the two conditions (control and test). As seen in Figure 2, height was underestimated in all conditions, including the control conditions. However, the pots were shorter when one of the modalities was reduced than when none of the modalities was reduced, F(1, 21) = 5.83, η2 = .22, p = .025. There was no significant effect of the modality that remained intact (p > .05), although numerically the shortening was less (i.e., the effect of sensory reduction was smaller) when auditory feedback was reduced than when haptic feedback was reduced and when visual feedback was reduced.
Results of Experiment 1. Left panel bottom to top: average height when haptic input and vision were available but audition was reduced, when audition and vision were available but haptic input was reduced, and when audition and haptic input were available but vision was reduced. Gray bars show heights when full input was available. Black bars show heights in the reduced input conditions. Right panel: percent height underestimation for the three groups in the three experimental conditions. AH = auditory and haptic reduced; AV = auditory reduced and visual reduced; HV = haptic reduced and visual reduced.
Years of experience had a positive relation with height when auditory input was reduced (r = .56, p < .05). Within this condition, the more expert the potters, the taller the pot (and so, the closer the pot height was to 5 in.). No other significant correlations were found.
Discussion
The results of the first experiment were consistent with the control hypothesis but inconsistent with the compensation hypothesis. Contrary to the compensation hypothesis, our participants did not make taller pots when they had less feedback. Instead, consistent with the control hypothesis, they made shorter pots when they had less feedback. The data hinted at the possibility that reduction of auditory feedback was least disruptive, as one might expect if one thought that seeing and feeling the clay was more informative than hearing the whirring of the wheel and the other sounds in the studio. This interpretation was corroborated by the fact that only in the auditory-reduced condition was there a relation between expertise and pot height. The more experienced the potters were, the less disturbed they were by the reduction of auditory input. Another way of saying this is that the potters who were less experienced were disturbed by the reduction of any kind of feedback of the kinds we varied, suggesting that they were not yet focused on feedback of a particular kind or kinds, whereas the more experienced potters had come to depend specifically on visual and haptic feedback and so were less disturbed by the reduction of auditory feedback. The idea that performers tend to become more and more reliant on particular forms of feedback as they become more and more proficient has been found in other domains as well (Proteau, Marteniuk, Girouard, & Dugas, 1987).
Experiment 2
In the second experiment, we did two things that we did not do in Experiment 1. One was to reduce input to two modalities rather than one. Our aims with respect to this aspect of Experiment 2 were as follows. First, we sought to determine whether we would replicate the shortening of the pots when sensory feedback was reduced. Second, we sought to determine whether the differences between modalities, such as they were in Experiment 1, would carry over into Experiment 2. Third and finally, we sought to determine whether the shortening of the pots, if it occurred in the second experiment, would be greater when two modalities were reduced than when one modality was reduced.
The other thing we did in Experiment 2 was to analyze the data with respect to a feature we did not analyze in Experiment 1, namely, the diameters of the pots as well as their heights; see Figure 3. As mentioned earlier, we gave no instructions about pot diameter. However, when experienced potters work with 1.5 pounds of clay (the weight used in Experiments 1 and 2), they generally produce cylinders whose height–diameter ratios are close to 3:1. We thought that we would learn more about the compensation versus control hypotheses by analyzing the produced ratios in our study. (Practical limitations made it impossible for us to record and analyze this aspect of the data in the first experiment.) We reasoned that if the compensation hypothesis were correct, reduction of feedback would cause the ratio of pot heights to pot diameters to exceed 3:1. The rationale for this prediction was that this larger ratio would reflect a subjective accentuation of the pot heights relative to their widths. By contrast, we reasoned that if the control hypothesis were correct, reduction of feedback would cause the ratio of pot heights to pot diameters to fall below 3:1. The rationale for this prediction was that making the pots short and stout would reflect a desire to increase the controllability of the still-developing pots. With a wider base of support, we reasoned, the centrifugal forces would be lower and so, like the proverbial teapot of the children’s song (“I’m a little teapot . . .”), it would be easier to manage.
Compensation and control hypotheses for the ratio of height to diameter. The dark gray region corresponds to the compensation hypothesis. The light gray region corresponds to the control hypothesis.
Method
Nineteen female and five male Northern Michigan University students (mean age = 24 years, standard deviation = 10.6 years) participated in the second experiment for a total of 24 participants. The method and materials were the same as in Experiment 1 except that participants had two reduced input channels rather than one. Sixteen participants were beginner potters (less than a year experience), six were intermediate potters (less than 5 years’ experience), and the remaining two were experts. We collected information about experience after the experiment was completed and confirmed that the distribution of experience in the groups was random. The number of participants we tested was again limited by the number of pottery wheels in the classroom.
The height and diameters of the produced pots were measured using a ruler, and pictures were taken for each pot. As for Experiment 1, we waited 2 months after the students’ enrollment in the course to ensure that beginner potters were comfortable enough with the task, as determined by their ceramics professor who is the fourth author of this article.
Results
Similar to Experiment 1, there were no statistically significant differences between the group means for the test condition, as determined by one-way ANOVAs for the completion times, F(2, 23) = 2.16, p = .14; final diameters, F(2, 23) = .16, p = .85; and final heights, F(2, 23) = .93, p = .41. For the control condition, the one-way ANOVA results between the three groups were also nonsignificant for time, F(2, 23) = .14, p = .87; height, F(2, 23) = .14, p = .87; and diameter, F(2, 23) = .14, p = .87. As in Experiment 1, participants produced vessels that were shorter than 5 in. The only statistically significant effect was related to the effect of removing any input (black bars in Figure 4) versus removing no input (gray bars in Figure 4), F(1, 21) = 15.21, η2 = .42, p = .001. The shortening, when input was removed compared to when it was not, was somewhat greater when auditory and visual input were reduced or removed (19%) than when auditory and haptic input were reduced (12%), and when haptic input was reduced and visual input was removed (11%), but the effect of reduced modality combination (or single modality preserved) was not statistically significant (p > .05).
Results of Experiment 2. Left panel bottom to top: average height when vision was available but auditory and haptic input were reduced, when haptic input was fully available but auditory and visual input were reduced, and when auditory input was fully available but haptic and visual input were reduced, for the three groups of subjects who were tested in the reduced input conditions (black bars) and in the control conditions, where full input was available for all modalities (gray bars). Right panel: percent height underestimation for the three groups in the three experimental conditions. AH = auditory and haptic reduced; AV = auditory reduced and visual reduced; HV = haptic reduced and visual reduced.
With respect to determining whether the reduction of two modalities resulted in more shortening than reduction of one modality, a one-way ANOVA showed that this was the case. The overall shortening was greater in Experiment 2 than in Experiment 1, F(5, 47) = 4.17, η2 = .31, p = .004. Post hoc Dunnett tests revealed a significant difference between the auditory-removal condition of Experiment 1 and the auditory-plus-haptic removal condition of Experiment 2 (p = .04) as well as the auditory-plus-visual removal condition of Experiment 2 (p = .01). When the auditory modality alone was reduced in Experiment 1, participants made taller vessels (4.55 in.) than when the auditory modality and visual modality were reduced in Experiment 2 (3.44 in.) and also when the auditory modality and haptic modality were reduced in Experiment 2 (3.54 in.). No other differences were statistically significant.
Potters’ experience had strong positive correlation (r = .70, p < .05) with height and also had a strong positive correlation (r = .79) with the height–diameter ratio in the auditory-plus-haptic removal (AH) condition. The more experienced the potters, the more the height was close to 5 in. and the closer the height–diameter ratio was to 3:1. Figure 5 shows the height–diameter ratio for all three conditions. There was a positive correlation (r = .77, p < .05) between potters’ experience and the height–diameter ratio they produced for the haptic-plus-vision removal (HV) condition. The ratio values varied between 0.84 and 1.90, and the more experienced the potters, the higher the ratio. Finally, a positive correlation (r = .77, p < .05) between experience and height–diameter ratio was found for the auditory-plus-visual removal condition, where the ratio varied between 0.71 and 1.48.
2
Height–diameter ratio for the three groups (AH = auditory and haptic reduced, AV = auditory reduced and visual reduced, and HV = haptic reduced and visual reduced).
Discussion
The results of the second experiment confirmed that reduction of feedback resulted in shortening rather than lengthening of the pots. Reducing sensory feedback in two modalities (Experiment 2) led to more overall shortening than reducing sensory feedback in just one (Experiment 1). Reducing auditory feedback led to less shortening in Experiment 1 than did combined reduction of auditory and visual, or auditory and haptic feedback in Experiment 2. The more experienced the potters, the taller their pots, consistent with the hypothesis that they had more control over the pots. Interestingly, the height–diameter ratios for all groups had the tendency to be smaller than 3:1, consistent with the control hypothesis. No matter which two modalities were reduced, the height-to-diameter grew with years of experience.
General Discussion
In this study, we explored the roles that auditory, haptic, and visual feedback play in the process of throwing clay pots. Studying this topic was interesting because we wanted to better understand the perceptual underpinnings of the skill involved in pottery. We began with the observation that potters get visual, tactile, and auditory feedback. (Smell may also play a role, but one we did not investigate.) A priori, one could imagine that potters come to rely on just one of these modalities or, in keeping with a growing body of research on multisensory integration (Rosenblum, 2010), they come to rely on more than one of them. Some potters can, of course, engage in their craft without the full array of modalities, for there are skilled potters who are deaf (Groth, Mäkelä, & Seitamaa-Hakkarainen, 2012) and others who are blind (https://www.youtube.com/watch?v=UXbu24e6120). It may also be that there are potters whose tactile sensitivity is impaired from potting itself because sensory neuropathy can be caused by sustained exposure to lead from ceramics and glazes (Sharma, 2009). Nevertheless, our findings are consistent with previous research that explore bimodal versus unimodal situations in perceptual texture judgments (Binns, 1937; Heller, 1982) where reduction of either modality impaired performance.
Given this backdrop, it seemed plausible that the potters in our study would be reliant on some modalities more than others, though it was unclear whether they would be more affected by the reduction of more than one form of feedback than just one. Our data showed that they were more affected by the reduction of more than one form of feedback than by the reduction of just one. Moreover, the form of feedback whose reduction had the smallest effect was the one that one might expect: audition. At the same time, our potters were not much more strongly affected by the reduction of touch alone or by the reduction of sight alone than by the reduction of hearing alone. It may be that losing any kind of feedback led to a general sense of loss of control that caused the potters to terminate their pot building sooner (at shorter heights) rather than later (at higher heights). 3
It is also possible that when haptic and visual feedback were removed, the sound coming from the wheel provided cues that were relied on more extensively than was typically the case. Consistent with this hypothesis, when auditory and visual feedback were removed, the potters bent down over the wheel to a degree not normally observed (Figure 6, left panel), as if they were trying to get a better listen to the sounds coming from the wheel. Similarly, when auditory and haptic feedback were removed, the potters again bent down over the wheel further as well (Figure 6, middle panel), as if they were trying to get a better look at the pot. Finally, when haptic and visual feedback were removed, the potters did not bend down any more than when they had full feedback (Figure 6, right panel), presumably because bending down further would not have boosted their feedback. All of these behaviors are consistent with the hypothesis that the potters were highly attuned to the feedback they normally received and adjusted their postures to boost the input that was still available. Relying on learned postures and being able to adjust those postures adaptively to achieve physical skills is likely to be a core feature of perceptual-motor control (Rosenbaum, Meulenbroek, Vaughan, & Jansen, 2001; Rosenbaum, 2017).
Potters bending over the wheel when deprived of auditory and visual feedback (left panel), when deprived of auditory and haptic feedback (middle panel), and when deprived of haptic and visual feedback (right panel).
This study, like any, had limitations. Five limitations can be identified. First, our sample size was smaller than might have been ideal. We were limited by the number of wheels in the classroom studio where this work was completed and by the number of semesters we were willing to wait before sharing our results. As mentioned in the Results section of Experiment 2, we had relatively low statistical power. Additional experimental situations and data could be used to check our findings.
Second, it is possible that our results stemmed from a perceptual illusion. Because our participants could not use a ruler 4 to measure the pots they produced, they had to rely on perceptual judgment about when the pots were the right size. It is possible that the shortening we observed in the reduced feedback conditions created or contributed to a height distortion illusion similar to the vertical–horizontal illusion (e.g., Wolfe, Maloney, & Tam, 2005) or the picture–surface illusion (Kennedy & Hammad, 2014).
Third, it is possible that our potters relied on the thickness of the pots rather than, or more than, the heights per se to terminate their pot building. Nothing in our data rule out this possibility and we are open to it, not least because wall thickness might be a determinant of the controllability of the emerging pot. A pot with thin walls is flimsier and therefore more susceptible to effects of stochastic variation in the pressures exerted upon it. Insofar as tall pots tend to have thinner walls than short pots, it is possible that wall thickness played an important rule in determining the stopping rule.
Fourth, we have, at best, a rudimentary understanding of the role of expertise. The expert potters were less thrown off by the reduction of feedback than were the less expert potters, and this difference was especially strong in connection with the reduction of auditory feedback. The nonexperts were more thrown off by the reduction of sound than were the experts. This could be because the nonexperts were more open to any kind of feedback they received, in which case reduction of feedback of any kind was disruptive. Alternatively, the experts may have had more strategies for coping with changes in feedback. For example, the experts may have previously found themselves in situations where the noise level in the ceramics studio suddenly went up and they had learned to lower their ears to hear the wheel better. By this account, the experts were better able to use compensatory strategies than were the nonexperts, which is perhaps why the experts make bigger pots than the nonexperts did. All of these are speculative remarks, however. Future research is needed to better understand the role of expertise in coping with reduction of sensory input.
Fifth, the gloves our participants wore reduced tactile sensitivity but did not alter proprioceptive feedback to any appreciable degree. Such feedback is likely to be important. Felt forces are likely to play an important role in building clay pots and, a propos the opening comments in this article about the steps involved in building a clay pot, they are likely to play a more critical role in some steps than others. Resistance force from the wheel provides information about the clay position relatively to the wheel but is likely to play a negligible role in the fourth and last steps. Future research can be done to test this hypothesis. In addition to asking potters to wear gloves, proprioceptive feedback can be modified through other means such as applying vibration to the cubital fossa (elbow pit) to observe its effects on the height and height–diameter ratio (see Goodwin, McCloskey, & Matthews, 1972).
These limitations notwithstanding, we think our results are quite clear in showing that the potters we studied conformed more to what was predicted by the control hypothesis than the compensation hypothesis. Rather than elongating the pots to offset the reduction of sensory input, as predicted by the compensation hypothesis, our potters made the pots even smaller, as predicted by the control hypothesis. For the control hypothesis to have been supported, it is reasonable to surmise that, when full sensory input was available, the potters relied on all of it, or each channel of it, to hone their pot shaping. To the extent that this interpretation is correct, it suggests that an essential part of the craft of forming clay pots is forming perceptual reference conditions (perceptual goal states) that enable potters to opportunistically sample the sensory feedback they receive in order to judge their progress as well as when to stop.
Footnotes
Acknowledgements
The authors thank Cheryl Konieczky for being the inspiration for this study and Morton Heller as well as three anonymous reviewers for extremely helpful editorial feedback.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported in part by a Committee on Research grant from the University of California, Riverside (to D. R.).