Figure 1
The 2D representation of a spherical image capture of a dormitory room (printed with permission from participant).
Table 1
Mixed-effects models for pointing error and pointing latency (log-transformed).
| PREDICTOR | POINTING ERROR | POINTING LATENCY | ||||||
|---|---|---|---|---|---|---|---|---|
| R2 = .32 | R2 = .30 | |||||||
| B | SE | t | p | B | SE | t | p | |
| Intercept | 19.30 | 1.30 | 14.88 | <.001 | 8.91 | 0.03 | 64.82 | <.001 |
| environment: familiar vs. unfamiliar | –0.60 | 2.59 | –0.23 | 0.82 | 0.11 | 0.07 | 1.60 | 0.11 |
| perspective: 0 vs. 45 | 11.88 | 1.38 | 8.63 | <.001 | 0.27 | 0.04 | 7.54 | 0.00 |
| perspective: 0 vs. 90 | –1.84 | 1.13 | –1.63 | 0.11 | 0.05 | 0.03 | 1.66 | 0.10 |
| perspective: 0 vs. 135/225 | 10.03 | 1.23 | 8.18 | <.001 | 0.23 | 0.03 | 6.65 | 0.00 |
| perspective: 0 vs. 180 | 0.21 | 1.14 | 0.18 | 0.86 | –0.02 | 0.03 | –0.57 | 0.57 |
| perspective: 0 vs. 270 | 0.93 | 1.00 | 0.93 | 0.35 | 0.07 | 0.03 | 2.81 | 0.01 |
| perspective: 0 vs. 315 | 11.27 | 1.48 | 7.59 | <.001 | 0.23 | 0.03 | 7.68 | <.001 |
| environment: familiar vs. unfamiliar * perspective: 0 vs. 45 | 2.33 | 2.75 | 0.85 | 0.40 | –0.03 | 0.07 | –0.37 | 0.71 |
| environment: familiar vs. unfamiliar * perspective: 0 vs. 90 | 3.76 | 2.26 | 1.66 | 0.10 | 0.06 | 0.06 | 0.97 | 0.33 |
| environment: familiar vs. unfamiliar * perspective: 0 vs. 135/225 | 0.15 | 2.45 | 0.06 | 0.95 | 0.07 | 0.07 | 1.06 | 0.29 |
| environment: familiar vs. unfamiliar * perspective: 0 vs. 180 | 3.04 | 2.29 | 1.33 | 0.19 | 0.05 | 0.07 | 0.72 | 0.47 |
| environment: familiar vs. unfamiliar * perspective: 0 vs. 270 | 6.21 | 2.00 | 3.11 | <.001 | 0.06 | 0.05 | 1.26 | 0.21 |
| environment: familiar vs. unfamiliar * perspective: 0 vs. 315 | 2.24 | 2.97 | 0.75 | 0.45 | –0.01 | 0.06 | –0.23 | 0.82 |
[i] Note: Each dependent measure is modelled as a function of the centered and contrast coded predictors: environment (familiar = –0.5, unfamiliar = 0.5), imagined perspective (with levels 0°, 45°, 90°, 135°/225°, 180°, 270°, defined as a single contrast with 0° as a reference category) and their interaction, using the maximal random effect structure possible. For each fixed effects and interaction, we report the unstandardized coefficient and its standard error, along with the associated t-value and p-value, and the overall variance captured by the model (R2). Statistically significant predictors (at the alpha = .05 level) are in bold.
Figure 2
a. Pointing error as a function of environment and imagined perspective. Error bars represent standard errors of the mean. b. Pointing latency as a function of environment and imagined perspective. Error bars represent standard errors of the mean.
Table 2
Mixed-effects models for pointing error and pointing latency (log-transformed) for participants in the Familiar environment condition.
| POINTING ERROR | POINTING LATENCY | |||||||
|---|---|---|---|---|---|---|---|---|
| FOR PARTICIPANTS WHOSE PREFERRED PERSPECTIVE IS THE SAME AS STARTING | ||||||||
| PREDICTOR | R2 = .22 | R2 = .23 | ||||||
| B | SE | t | p | B | SE | t | p | |
| Intercept | 9.15 | 1.74 | 5.27 | <.001 | 8.57 | 0.08 | 113.09 | <.001 |
| perspective: Other Orthogonal vs. Preferred/Starting | 1.14 | 1.95 | 0.59 | 0.57 | 0.19 | 0.08 | 2.38 | 0.04 |
| FOR PARTICIPANTS WHOSE PREFERRED PERSPECTIVE IS DIFFERENT THAN STARTING | ||||||||
| PREDICTOR | R2 = .22 | R2 = .31 | ||||||
| B | SE | t | p | B | SE | t | p | |
| Intercept | 16.60 | 2.14 | 7.77 | <.001 | 8.82 | 0.07 | 130.23 | <.001 |
| perspective: Starting vs. Other Orthogonal | 3.26 | 1.39 | 2.34 | 0.02 | 0.08 | 0.04 | 1.83 | 0.08 |
| perspective: Starting vs. Preferred | 5.23 | 1.66 | 3.15 | <.001 | 0.07 | 0.05 | 1.48 | 0.15 |
[i] Note: Each dependent measure is modelled as a function of the centered and contrast coded predictors, using the maximal random effect structure possible. For participants reported preferred orientation was the same as the starting orientation (N = 11), perspective type was contrast coded as: preferred/starting = –.5, other orthogonal = .5. For participants reported preferred orientation was different than starting orientation (N = 21), perspective type was contrast coded with a simple contrast with “starting” as the reference category. For each fixed effect, we report the unstandardized coefficient and its standard error, along with the associated t-value and p-value, and the overall variance captured by the model (R2). Statistically significant fixed effects (at the alpha = .05 level) are in bold.
Figure 3
Violin plots representing the distributions of the pointing performance of participants in the Familiar environment condition.
a. and b. illustrate the pointing error and latency of participants whose preferred perspective was the same as the starting perspective (N = 11), according to perspective type (preferred/starting vs. other orthogonal perspectives). c. and d. illustrate the pointing error and latency of participants whose preferred perspective was different from the starting perspective (N = 21), according to perspective type (preferred vs. starting vs. other orthogonal perspectives). Boxplots represent the median and quartiles (Q1, Q3), and the red dot represents the mean of each condition. Black dots indicate observations with values greater than Q3 plus 1.5 times the interquartile range.
Table 3
Mixed-effects models for pointing error and pointing latency (log-transformed) for participants in the Unfamiliar environment condition.
| POINTING ERROR | POINTING LATENCY | |||||||
|---|---|---|---|---|---|---|---|---|
| FOR PARTICIPANTS WHOSE PREFERRED PERSPECTIVE IS THE SAME AS STARTING | ||||||||
| PREDICTOR | R2 = .13 | R2 = 17 | ||||||
| B | SE | t | p | B | SE | t | p | |
| Intercept | 13.38 | 1.57 | 8.51 | <.001 | 8.86 | 0.06 | 44.64 | <.001 |
| perspective: Other Orthogonal vs. Preferred/Starting | 0.36 | 1.41 | 0.25 | 0.80 | 0.16 | 0.04 | 3.91 | <.001 |
| FOR PARTICIPANTS WHOSE PREFERRED PERSPECTIVE IS DIFFERENT THAN STARTING | ||||||||
| PREDICTOR | R2 = .33 | R2 = .10 | ||||||
| B | SE | t | p | B | SE | t | p | |
| Intercept | 16.59 | 3.87 | 4.29 | <.001 | 8.84 | 0.05 | 195.61 | <.001 |
| perspective: Starting vs. Other Orthogonal | –0.81 | 2.35 | –0.35 | 0.73 | 0.04 | 0.06 | 0.66 | 0.52 |
| perspective: Starting vs. Preferred | 0.48 | 2.62 | 0.18 | 0.86 | –0.03 | 0.07 | –0.41 | 0.69 |
[i] Note: Each dependent measure is modelled as a function of the centered and contrast coded predictors, using the maximal random effect structure possible. For participants reported preferred orientation was the same as the starting orientation (N = 21), perspective type was contrast coded as: preferred/starting = –.5, other orthogonal = .5. For participants reported preferred orientation was different than starting orientation (N = 11), perspective type was contrast coded with a simple contrast with “starting” as the reference category. For each fixed effect, we report the unstandardized coefficient and its standard error, along with the associated t-value and p-value, and the overall variance captured by the model (R2). Statistically significant fixed effects (at the alpha = .05 level) are in bold.
Figure 4
Violin plots representing the distributions of the pointing performance of participants in the Unfamiliar environment condition.
a. and b. illustrate the pointing error and latency of participants whose preferred perspective was the same as the starting perspective (N = 21), according to perspective type (preferred/starting vs. other orthogonal perspectives). c. and d. illustrate the pointing error and latency of participants whose preferred perspective was different from the starting perspective (N = 11), according to perspective type (preferred vs. starting vs. other orthogonal perspectives). Boxplots represent the median and quartiles (Q1, Q3), and the red dot represents the mean of each condition. Black dots indicate observations with values greater than Q3 plus 1.5 times the interquartile range.