mmm... i don't know, you need to find some documentation, because who know how is this thing on the inside. It uses a positive supply to power the meter. The yellow wire seems to be to measure the voltage, and then the thick red/black wires to measure current (well, black probably for voltage too).
First, measure both black wires to make sure that they aren't shorted inside (probably won't, but just to be sure...). Then, on the points labeled "power supply for the load", you could connect GND to + and VEE to - to measure your PSU's negative output, or connect VCC to + and GND to - to measure your PSU's positive output.
(VCC means your PSU's positive output wire and VEE your PSU's negative output wire).
Well, at least that's what makes sense to me, but no warranties without documentation.
I think that you need to switch the signals on "power supply for the load", because even if the ammeter works in any polarity, there's only the yellow cable to measure voltage, and that means that the other cable for the voltmeter is probably the black one for the ammeter, so you cannot just leave the ammeter wired directly.
And that's not all, you also need to connect your PSU's output directly when the meter is connected to the other side, so you have to switch VCC, GND and VEE on two different possitions, further complicated by the fact that the meter needs to be in series with the load to measure current. Worse: you have only one GND terminal, and it's shown that the ammeter goes in series with the load, over the black wire. If that's the only way to use it, well, that's a problem, because it's a split supply, so that the current from both sides will return on that lead (but going in different directions).
That seems like several things to test, then draw how the meter would need to be connected to measure the positive and negative output, then see how a switch can provide that kind of connections. And even then, moving the switch will briefly interrupt the output.