I know this is slightly off-topic, but the study referenced for the first graph [1] seems absurdly badly designed. Subjects were divided into arbitrary age grou

I know this is slightly off-topic, but the study referenced for the first graph [1] seems absurdly badly designed. Subjects were divided into arbitrary age groups, but all subjects were given the same task, so why divide at all? Why not give a scatter plot of age vs. perceived seconds? It's so unnecessary -- it's basically just to get an ANOVA -- that it makes me think it is p-hacking to try to find significance rather than actually measuring anything of interest. [1] https://www.scielo.br/j/anp/a/d6SvJK5tM6kCFPTmpVj5pSz/?format=pdf&lang=en https://www.scielo.br/j/anp/a/d6SvJK5tM6kCFPTmpVj5pSz/?forma...