Interpreting pairwise comparisons for significance when you have 3 groups

Dear posters,

I am currently trying to interpret results from three different groups in my study. There is essentially a control group and two picture groups. My original hypothesis would be that one of the picture groups would outperform the other, but post hoc testing revealed something else.

In absolute terms, the participants in the no picture group (control group) scored the lowest on delayed posttests. The picture group I expected to perform less well outperformed the no picture in absolute terms (but not sig.) but performed less well than the hypothesized superior picture group (but again not sig.). However--there is a significant difference between the no picture and the superior picture group.

Even though there was significance at each end of these sort of ordinal categorical conditions (no picture, weak picture, strong picture), there was no effect found between pictures or between weak picture and no picture. Can I still claim an effect for the strong picture group since it was the only one that significantly outperformed the no picture group even though it didn't outperform the weaker picture group?

Thank you for any suggestions you can provide! I am fairly new to statistics and any input you can provide would be much appreciated.