Two aims were pursued with the study reported here, first the developm
ent of training guide-lines to improved different subcomponents of emo
tional imagery, and secondly the examination of effects on different c
omponent measures after realization of the training in different sampl
es. The development of the training guide-lines was based on Lang's (1
979) bio-informational theory of emotional imagery and subsequent rela
ted studies. It is proposed that relaxation, vividness and controllabi
lity of imagination interact in influencing the ability to emotional i
magery and that these components can be improved by training. Training
exercises for each of these components were constructed or adapted fr
om existing imagery interventions and were described in a manual. Two
studies were conducted to test the assumption that the manual-based tr
aining effects the assumed subcomponents and improves emotional imager
y. Dependent measures consisted of imaginative competence questionnair
es, subjective change ratings and ratings related to the evaluation of
the training. In addition mood and bodily sensation ratings were asse
ssed before and after the provocation of autobiographical recollection
s of different emotional qualities. Results in both study samples supp
orted the effectiveness of the training in improving emotional imagery
. The ambiguous findings of the autobiographical recollection test wer
e discussed in relation to methodological difficulties, and conceptual
and methodological changes were suggested. It was concluded that the
studies represent a first step to clinically useful interventions aime
d at improving subcomponents of imagery which are seen as prerequisite
s of many important therapies.