Seems like an unlikely (and therefore interesting) finding.
If you could export the data in a CSV format then it would be easier to analyse any correlation. Just looking at graphs of averages can be misleading — human brains are tuned to notice patterns, after all, even if they're not really there.
Importantly, even if the two are correlated, they might not be causally related. It could just be a coincidence. It's also possible that some third factor is causal in both of them. E.g. time of year or diet.
If I were to hypothesise a connection between the two, then perhaps when you feel good about your body (because you've lost weight) you psychologically predispose yourself to faster reactions. Indeed, we suspect the inverse of that can occur -- students primed with words associated with old age walked out of the experiment more slowly than the control group.
But that's just conjecture on my part.
Keep us posted on any further analysis!