You want to test if there is a significant change in the frequency of outliers. I suggest trying a Poisson loglinear model.
Note: if the size of the sample for each condition varies, then you must create an offset for the model. This must be the log of the sample size.
Select Analyze > Fit Model. Select the data column with the outlier indicators and click Y. Select the data column with the condition indicators and click Add. Click Standard Least Squares > Generalized Linear Models. Click Distribution > Poisson. Use the default link function (log). If you need to use an offset, select this data column and click Offset. Select the option for Overdispersion. Then click Run.
The results should help you determine if there is a significant difference between the conditions.
Just to add a bit to my colleague @markbailey's advice and counsel which makes perfect sense...I also recommend just plotting the actual performance data using say Graph Builder as a starting point. Especially if you are looking for outliers...a picture can tell you alot that might be masked or not visually apparent using a modeling approach all by it's lonesome.
I was actually thinking of/ looking for something which could give me an indication of increased frequency of outliers upon doing my comparison between different test conditions for numerours parameters. For example, for mean compararison, my first gating creteria would probably be the T ratio and P value. Therefore, I was just thinking of having something similar which I could base on to monitor the outliers' frequency performance.