Fig 1.
Diagram illustrating the regions of the tongue tested.
Locations (L1 –L4) and Subarrays (S1–S4) were identified on tongue photographs by applying a 4 cm2 grid overly with Photoshop software. The blue box shows the regions of the tongue that were analyzed in this study. These regions were chosen because they demonstrated the highest sensitivity and discrimination ability. In addition, the tongue surface was the most likely to maintain contact with the electrode array during testing in these regions, thus reducing a potential source of error.
Fig 2.
Tongue maps demonstrating average reported perceived intensity for the tested region of two subjects.
The left map shows results from a participant with higher overall perceived intensity for ETS, with highest intensity reported for electrodes in anterior medial regions (Location 1, Subarrays 2 and 3). The right map shows results from a participant with lower perceived intensity across the tested area as indicated by the darker colors. The scale to the right of each map reflects the intensity represented by color and the orientation of the map is indicated by the labels surrounding each map. Note that for both participants, stimulation of the anterior region resulted in higher perceived intensity than stimulation of the posterior region.
Fig 3.
Box plots for perceived intensity and discrimination ability.
Left: average perceived intensity was significantly higher in Location 1 relative to Location 2. Right: average discrimination ability was only slightly better in Location 1 as measured by the average number of electrodes which were perceived across all participants. Better discrimination ability is indicated when the average is close to two since only data collected while two electrodes were active were used in this analysis.
Fig 4.
Tongue maps demonstrating the minimum discrimination distance for the tested region of two different subjects.
For these maps, orange indicates that the subject was able to discriminate two active electrodes that were 2 mm apart whereas blue indicates the participant was able to discriminate electrodes that were 6–10 mm apart. The left map shows results from a participant who was able to discriminate closely spaced electrodes on the anterior centimeter (L1) and on the right side of the tongue (S3). The right map shows results from a participant who had less discriminatory ability overall as indicated by more blue in the map. This participant had the best discrimination on the left side of the tongue (S2). These two maps illustrate the extreme variability between some individuals with respect to discrimination ability in the anterior medial tongue (L1 & L2, S2 & S3).
Fig 5.
Fungiform papillae (FP) counts from all participants.
A. The total number of FP in all 4 regions varied among the participants. Eighty percent of participants had between 51–150 papillae and 40% of participants had between 101–125 papillae. B. Average FP number in each Subarray. The most medial two subarrays in location 1 (L1S2 and L1S3) had the highest average number of papillae in the participant pool. C. Average FP number in location 1 was higher than location 2. D. The average FP number for Subarray 2 (left) and Subarray 3 (right) was 54 and 55 respectively, and not statistically significant.
Table 1.
Numbers of papillae for individual subjects to the left of the midline (S2) compared to the right of the midline (S3) for Locations 1 and 2 combined.
Fig 6.
Relationship between papillae density and perceived intensity.
Each dashed line represents the conditional subject-specific profile calculated by incorporating the best linear unbiased predictions of all the random effects for each subject. The red line is the marginal line from the fitted linear mixed model for a typical, average subject; that is nominal Taste (31.33) averaged over Location, Subarray, Distance and Orient. The 15 black dots are the observed means (n = 32) for each subject. Note that the red line is quite flat which indicates that on average, there is no association between papillae density and perceived intensity (P-value = 0.3759).
Table 2.
Tests of fixed effects for perceived intensity of ETS.
Fig 7.
Relationship between perceived intensity and fungiform papillae density for one participant relative to the average relationship in the set of participants.
The dashed blue line indicates the conditional subject-specific profile calculated by incorporating the best linear unbiased predictions of the random effects for this subject. The red line is the marginal line from the fitted linear mixed model for a typical, average subject as shown in Fig 6 and the black dots indicate the observed means. Blue Xs are the raw observed perceived intensity means at each Location-by-Subarray region of the tongue (n = 8) for one subject, while the blue diamond is their overall mean. The blue dashed line is a conditional line averaged over Location, Subarray, Distance and Orientation obtained from the fitted model. Notably, we found evidence of a (positive) linear association between papillae (Pap) and perceived intensity (PerInt, P value = 0.0361) for this participant.
Fig 8.
Overlay of perceived intensity map and fungiform papillae for the participant that demonstrated a statistically significant positive linear association between papillae and perceived intensity.
Colors represent perceived intensity (PerInt), with lighter blue and green indicating higher PerInt as indicated by the scale to the right of the tongue map. Black dots represent individual fungiform papillae (FP) for this participant in the related regions. Note that there are more FP in the anterior region where there is also higher perceived intensity.
Fig 9.
Relationship between papillae density and perceived number of active electrodes.
The red line is the marginal line from the fitted linear mixed model for a typical, average subject. The 15 black dots are the observed means (n = 32) for each subject. Note that the red line has a positive slope, which shows that there is an association between papillae density and perceived number of electrodes (P-value = 0.0203).
Table 3.
Type 3 tests of fixed effects for perceived number of electrodes.