In a comment, Coleman Ridge observes that whatever we do to help put Aikido principles to use in daily life situations, for effective results, we ought to build new practices, and not hope for success by simply "intending" to do something different. He conjectures that these new practices would be rhetorical or linguistic, basically arising in the realm of speaking and listening.
This is what I think too. An assumption I'm making is that practices aimed at producing new actions, different from our default programmed actions, will be most successful if we simulate training environments similar to the ones in which we actually want to move in new ways. What I mean is, if we want to verbally "blend" with a declared adversary at work, we would succeed best by designing and training with practices using linguistic "blends" staged or role-played in a workplace setting.
This assumption is a corollary of my original suspicion: that on-the-mat training is insufficient to enable a robust ability to enact Aikido principles verbally in off-the-mat domains. I do not know whether these assumptions are that well founded, and I believe it would be a tremendous project for a doctoral student to explore these assumptions.
I also have additional concerns about any proposed verbal learning practices. One of the most important of these concerns to me is, how to facilitate such learning while preserving the dignity of everyone in the loop. In the dojo this is easy. The students give the teacher complete discretion to guide them through whatever physical and attitudinal shifts they need to make to progress in the art. It is a given that the students are not themselves aware of what these shifts are nor how to make them. To put ourselves in the hands of a teacher to help us learn what we don't know we don't know requires a deep trust. Establishing this kind of trust in the more purely linguistic world of everyday life and especially work life is vastly more difficult and complicated than doing it in a dojo, where you can see the physical techniques you want to learn performed elegantly and expertly.
There have been a few researchers in organizational behavior that I have grown to respect as possible resources for help in designing the kinds of practices that might help, and at the same time, satisfy the learner's right to informed consent to the nature and structure of the learning process design. This value of informed consent, I feel, is important in preserving the dignity of everyone involved.
Some of the researchers I'm talking about are John Dewey, Kurt Lewin, and Chris Argyris. One basic practical guideline I'm drawing from these and other researchers is that it is not helpful to "trick" someone into learning something, no matter how beneficial such learning potentially may be to the student. Not only is it not helpful for creating learning, it actually inhibits learning by eroding the student's trust in the teacher or facilitator. I think many teachers limit their success by being in denial that their very methods are counterproductive to the results they claim to achieve.