Figure 1
The figure contains three vertically stacked error-bar graphs comparing two groups labeled “Chat G P T” and “Non-Chat G P T” on the horizontal axis. Panel (a) shows subject-specific standard deviation for lemon concentration. The vertical axis ranges from 0.00 to 2.00. The Chat G P T group has a mean of approximately 0.59 with an error range from about 0.47 to 0.71. The Non-Chat G P T group has a higher mean of approximately 0.97 with an error range from about 0.69 to 1.26. A Mann–Whitney U test indicates a statistically significant difference (p < 0.05). Panel (b) shows subject-specific standard deviation for sugar concentration, with the same vertical axis range. The Chat G P T group has a mean of approximately 0.95 (error range 0.73 to 1.18), while the Non-Chat G P T group has a mean of approximately 1.15 (error range 0.92 to 1.37). The Mann–Whitney U test indicates no statistically significant difference (p > 0.1). Panel (c) shows subject-specific standard deviation for price. The Chat G P T group has a mean of approximately 0.91 with an error range from about 0.72 to 1.11, while the Non-Chat G P T group has a mean of approximately 1.07 with an error range from about 0.85 to 1.28. The Mann–Whitney test again indicates no statistically significant difference (p > 0.1). All numerical values are approximate.rtically stacked error-bar graphs comparing two groups labeled “ChatGPT” and “Non-ChatGPT” on the horizontal axis. Panel (a) shows subject-specific standard deviation for lemon concentration. The vertical axis ranges from 0.00 to 2.00. The ChatGPT group has a mean of approximately 0.59 with an error range from about 0.47 to 0.71. The Non-ChatGPT group has a higher mean of approximately 0.97 with an error range from about 0.69 to 1.26. A Mann–Whitney U test indicates a statistically significant difference (p < 0.05). Panel (b) shows subject-specific standard deviation for sugar concentration, with the same vertical axis range. The ChatGPT group has a mean of approximately 0.95 (error range 0.73 to 1.18), while the Non-ChatGPT group has a mean of approximately 1.15 (error range 0.92 to 1.37). The Mann–Whitney U test indicates no statistically significant difference (p > 0.1). Panel (c) shows subject-specific standard deviation for price. The ChatGPT group has a mean of approximately 0.91 with an error range from about 0.72 to 1.11, while the Non-ChatGPT group has a mean of approximately 1.07 with an error range from about 0.85 to 1.28. The Mann–Whitney test again indicates no statistically significant difference (p > 0.1). All numerical values are approximate.Average innovation scores for continuous parameters. Notes: Bars represent mean values; error bars denote 95% confidence intervals. Mann–Whitney U test indicated a statistically significant difference between groups regarding lemon concentration (p-value < 0.05)