En los productos audiovisuales, el canal oral y el visual transmiten información simultáneamente. Si la coherencia de sentido entre texto e imagen desaparece en el doblaje, el objetivo de verosimilitud no se satisface. En este trabajo analizo y clasifico las incoherencias texto-imagen y aporto posibles soluciones a este recurrente problema.

When dubbing audio-visual products, translators must take into account the interactionbetween the oral and the visual channels, as both of them convey information at thesame time. It is sometimes difficult to maintain the original coherence between text andimage (due to wordplay, for instance), and the dubbed version ends up making no sense.This has negative consequences on the audience because they realise they are watchinga translation and that it could actually be wrong. Hence, the purpose of this paper is tofind out what causes the appearance of such incoherences and give some generalsolutions or strategies to prevent them. To do so, I have analysed specific examples oftext-image incoherences in the Spanish version of The Simpsons, as this animationseries often transmits information and humour through the visual channel. The results ofmy work are a classification of the different types of text-image incoherences (basedupon the several restrictions imposed by the image) and a reflection on the importanceof avoiding them to improve the quality of dubbed versions. In some cases I proposeadaptations of the text to solve incoherences, but in other examples visual restrictionsmake such adaptations almost impossible, which leads me to the conclusion that digitalmanipulation of the image is sometimes necessary. Future studies could broaden thescope of my work by analysing more examples of text-image incoherences in otherseries and films: this could help improve my classification of them and give furthersolutions to this recurrent problem.