Most definitely. Fixations are usually defined as the period between saccade offset and the next saccade onset, so vice-versa you can think of saccades as occurring between two fixations. It is worth noting that there is a disagreement about whether microsaccades occurrence define new fixation periods, and I see papers both using and disregarding microsaccades in that regard, however this is also likely due to technical limitations (i.e. too low sampling rate in the video-based eye-tracker used).

In our lab at Tel-Aviv University we use Engbert and Margenthaler's algorithm you referred to on a regular basis to examine saccade and microsaccade-rate, amplitude, velocity etc., it is quite the standard in the field so you can't go wrong with it.

Noam Tal-Perry Thank you for your response.

I know this is an old question, but...

num(saccades) does not always precisely = num(fixations), depending on a couple factors.

I can readily think of several situations where this would be the case:

(1) you may have fixations interrupted by blinks (or data loss!)... You fixate, then blink, and then you might label the next period as a new fixation (likely on a different part of the image) but there was no detected saccade per se.

(2) either the number may be off by +- 1 or 2 depending on where the trial ends. For example, you may have a fixation and then a saccade during which the trial ends and data analysis stops, or you may have a fixation and then a saccade and then a fixation at which point the trial ends and data analysis stops. You might start a trial period during a fixation or during a saccade, and you might end a trial period during either a fixation or a saccade.

(3) if you are omitting parts of your data from analysis for some reason (i.e., not considering eye position within a certain amount of time after a blink or saccade) you may end up removing some fixations or saccades.