This is why rapid switching between two native languages is the easiest way for coordinate bilinguals to talk (listen to Latin American folks on the NYC subway). No tip of the tongue there, and always the most apt expression at the right time.
That's really irrelevant to the point at hand (as well as rather speculative.
The disfluencies that this study looks at regards rarely used words. Fluent casual speech isn't the best place to look for such disfluencies anyway, so i don't know that listening to people on the subway is a good demonstration.
Also, while code-switching may be a way to get around tip of the tongue experiences, it is a considerably more complicated phenomenon, and there is likely to be a variety of other factors and considerations to take into account when studying code-switching.
The speculation isn't unfounded though, although it isn't appreciably strong here. Word choice comes substantially faster when the most efficient expression can be chosen without any self-monitoring. It's quite plausible that in this case you'll find less instances of tip of the tongue.
While the experiment here relied on rare words, and I'm just one data point, but I get ToT probably once for every 2000 words I say. When rapid code-switching is consciously available, I can quickly avert it because I have fast access to good candidate words, which reduces the apparent occurrence.
If ToT is taken as a speech disfluency from unresolved competition for word access, you could mitigate it simply by creating an imbalance between the signal strength such that winner takes all (in which case the subway example does demonstrate something useful, although as you point out, the variables are jumbled up). Somehow I got a deja vu just now. This has been discussed here before, or somewhere, or what I am writing is coming from something else I have read...
I also realize that ToT cannot be reduced to simple competition, but it is also plausible that slow access and competition manifest similarly, i.e. obstructed speech flow, while partial semantic access. WEAVER++ tried to explain this and I liked it very much (obviously I don't know what's state of the art now).
