Rapid E-learning - Is it more Behaviorist or Constructivist based?

I'd say neither most of the time. And situationally both or one or the other the rest of the time. It's driven by the wind of the design.

I say neither most of the time because much of the Rapid stuff I see (which isn't much different than the more expensive stuff I see) is simply an information-coaster with a quiz. A windblown cramming session that delivers and assesses short term recall. I suppose in some ways this could be considered behaviorist. But really we aren't doing anything to influence volition, other than the volition to click that close button when I score high enough to put the course behind me. Both of those theories target influencing behavior (either externally or internally). Most courses don't really have this explicit effect even if it's a stated goal. So I say neither.

To be behaviorist, I believe a course has to influence via compulsion (recognize, respond, good dog). I do think there are designs that specifically drive compulsion using these methods. This is often done through drills and repetition or psychological games that drive a compulsive behavior based on a simple set of stimuli or through rote repetition through remediative torture cycles. There are likely more standalone elearning courses that fit into this model loosely than a constructivist model, but even these tend to fall back on the info-coaster model.

Constructivist approaches are often a challenge for designers that target self-directed and packaged outputs. I saw this quite a bit in more social environments or facilitated study opportunities as presented in online courses taken through university. But we rarely see this in standalone eLearning courses.  This requires some pretty keen planning and systems to account for speed of uptake - adapting and adjusting the pathways or providing self-adjustment mechanisms for the learner. This might also require significant trust in the learner on the part of the designer and a willingness to let the learner venture beyond the training package into the outside world (in some cases the real world) to explore and retrieve feedback.

We're working on two courses that take on some behaviorist qualities and some constructivist qualities. The first course will have a drilling system for learning morse code for allied visual communications. This will consist of a virtual "keyer" for sending code as a part of the instructional sequence. This drill mechanism will allow the learner to respond to and send messages, providing feedback that includes their words per minute score (for mastery progression - this is the compulsion piece - strongly behaviorist but also a bit constructivist) and the actual message they sent as interpreted by the key interpreter (pushing it further into the constructivist realm as the learner can experiment with assessing their own skill level and make adjustments to their own performance).

The other courselet we're looking at is maneuvering boards (MOBOARDS). This is a relatively simple concept but takes practice and feedback and is ripe for a constructivist approach. One of the activities we're looking at implementing is a "learn to judge and score your own work" where we provide the opportunity to complete a board on paper using the real tools and follow up the exercise with a "Did your board look like this? (Y|N) -- OK, which of these is it closest to? (1|2|3)" --- "OK, here's where you went wrong. Let's try another very similar problem" The idea here is we aren't just training folks how to execute the task, we're trying to instill evaluation skills so they can easily "troubleshoot" their own performance an easily / naturally check their work.  And we really want to do this using objects and interactions in the real world. We trust the learner will want to do these things - and we believe that leveraging the one human we left in the equation is the only right way to go

Neither of these is a pure approach. Pure approaches are winged unicorns. They only exist in textbooks:P