Should electric fuel pump be ON at high altitudes?

The reason would to prevent pressure of the fuel supplying the engine dropping below its vapor pressure. However on my plane the boost pump is not a boost pump at all, it is actually an alternate pump that produces about the same pressure in the same lines as the mechanical pump. I have no idea if the situation varies on other types.

In my single experience with vapor lock the manifestation was the engine going quiet, unfortunately at around 800 ft AGL The solution was a boost pump to augment gravity feed where the fuel is warm, in front of the firewall.

You would notice a rough running engine.
The ambient pressure would be low and therefore the chance of vapour lock is increased.
But the temps would be quite low, so the chance of vapour lock decreases.

Choose your winning theory.
A friend with a Seneca has rough running engines at altitude sometimes and turns on his electric pump.

There is no one answer possible unless at the very least we know what type of aircraft is discussed.

Just for a single beginning: there’s a world of difference between (fuel tanks in the) low-wing vs. high-wing craft. The latter can do without any kind of fuel pump under certain conditions, thanks to the never failing force of gravity; the former had better have an electrical pump to back up the (hopefully!) already present mechanical fuel pump.

Not to even begin discussing diesel or jet fuelled planes.

A continental O-200 just smoothly loses revs and stops with vapour lock – on take-off run with mogas. I had checked the fuel tank temperature, but heat from the engine when parked into wind had heated the gascolator and piping. The engine restarted after a few attempts, and gave taxi revs to clear runway. When the gascolator was drained, immediately after clearing the runway, froth came out, confirming vapour lock. The electric pump was on for take-off, and there was some gravity feed from the front tank – but full power produced vapour lock. It was very similar to fuel off – just fuel starvation..
With a 3.4 litre deisel on a boat, I’ve had vapour lock with cold deisel, showing as vapour in the clear top of the filter, causing loss of revs but not stoppage. On shut down, the bubbles vanished. The cause was a partial blockage in the line from the fuel tank.
If you had a partial blockage tank-side of the pumps, reducing power would help, but more pumping would not.

I am thinking about a low wing aircraft with fuel injection. All these must have a fuel pump…

They do. An engine driven mechanical fuel pump. Most of these are perfectly capable to pump/pressurise a mixture of fuel, air and vapor. The electric fuel pump is only required when the engine driven pump fails. To prevent the engine from surging during the time it takes from recognition of the failure of the engine driven pump to manual selection of the electrical pump, the latter shall be switched on during critical phases of flight (takeoff, landing and airwork). If the manual of the aircraft doesn’t specifically require the electrical pump to be on under certain conditions (e.g. high altitude), I would rather leave it off. Because the basic principle for aircraft systems is always the same: “The less you use it, the longer it lasts.” And you never know when you will need it in anger!

I cannot imagine there would be a bubble of AIR – air should never be able to enter the fuel lines. Or there should be a leak somewhere, dripping out fuel when operative, sucking in air when not. Do watch for fuel leaks!

Vapour lock is caused by a bubble of vapourised fuel that cannot get away. In my own bird I relaid the fuel lines so that they form one continuously rising line from the header tank through the panel mounted fuel cock, through the miniature fuel filter, to the mechanical pump on the engine, exactly to counter vapour locking: any vapour that does form will raise to the fuel pump and be pushed to the carbs.

Causes of vapour locking are thus (both must occur):
1) formation of vapour, caused either by high temperature (no fuel lines close to body plating, where they will heat when parked in the sun) or by low pressure, as in the mentioned story of a (partial) blockage before the pump;
2) presence of a non-vented high point where a bubble of vapour can get blocked – fuel lining should be one continuously rising line, but perhaps this is not always easy to realise.

(I hope I have not only stated the obvious, but I can’t see anything more to say about it)

PS in my world of microlight flying, it is far from obvious to have a panel full of dials. Mine carries Rpm, CHT1, Oil temp and Oil pressure, and that’s all for the engine. The guru, who is a strong advocate of keeping things as rudimentary as possible, has once reluctantly said that if any instrument is worth adding, it is a fuel pressure gauge. Though I am not sure if fuel pressure is affected by the formation/presence of fuel vapour, actually I don’t think so.

The reason for electrical back up pump for injectected engines is mainly, that mechanical pump on engine has lower and lower pressure because of altitude.
Mechanical pump elements are spring and membrane with two one way valves and link to lift the membrane against the spring. The forces to give the fuel pressure are the springload wich is steady and the pressure in crankcase on springside of membrane wich is not steady but dimnishes with increasing altitude.
The electrical pump is needed when the injectors dont fully atomice the fuel and droplets are reason for engine running ruff.
For example with full power my RV has almoust same pressure mechanical/ electrical on sealevel 26/28psi but on FL50 they are 23/27