Hi. Yes, with a code component, you can maintain a running total of RTs for each trial type and then take their mean at the appropriate time for feedback. Probably easiest if you show us the variable names in your conditions file, but let’s say you have a column for letter colour and a column for the text content.

Put something like this in the `Begin Experiment`

tab:

```
# initialise some variables:
n_congruent = 0
rt_congruent = 0.0
n_incongruent = 0
rt_incongruent = 0.0
```

Something like this in the `End Routine`

tab of your experimental routine:

```
if letter_colour = text_content: # use your actual conditions file variables
n_congruent = n_congruent + 1
total_rt_congruent = rt_congruent + your_keyboard_comnponent.rt
else:
n_incongruent = n_incongruent + 1
total_rt_incongruent = rt_incongruent + your_keyboard_comnponent.rt
```

And something like this in the `Begin Routine`

tab of your feedback routine:

```
mean_congruent = total_rt_congruent / n_congruent
mean_incongruent = total_rt_incongruent / n_incongruent
```

Those variables can then be used in a text component to give feedback.