Do Driving World-Action Models Imagine What They Plan? Image–Action Consistency for Evaluation and Learning

Driving world-action models (WAMs) pair a native action with an imagined future. Interpreting these outputs as one coherent prediction requires evidence that they describe compatible motion and respond together to changes in conditioning. We introduce Image-Action Consistency (IAC), which evaluates this relationship through candidate-blind visual measurement and explicit observation coverage. IAC maps video and action into a shared space of yaw and interval longitudinal progress. The Alignment Score (AS) measures within-pair motion agreement, the Response Consistency Score (RCS) checks paired yaw responses to controlled condition changes, and the Grounding Score (GS) measures geometric agreement with independently logged futures. On held-out logged videos, visual progress estimates achieve $90.1\%$ tolerance accuracy on supported intervals, with $88.2\%$ coverage. Under the tested DriveWAM intervention protocol, response directions agree in $92.5\%$ of pairs with detectable responses in both outputs, yet only $49.5\%$ of all $745$ valid pairs produce material action responses. This gap shows why conditional agreement must be interpreted together with response coverage. Paired visual-motion targets also improve PDMS over matched-budget action-only fine-tuning by $1.54$ points on DriveWAM and $1.00$ point on WorldDrive. IAC thus provides both a diagnostic of joint predictions and a source of motion supervision for action learning.