regress command with a binary predictor

I am trying to see whether being rewarded '1' or not rewarded '0' in trial 't' predicts latency of responding in trial 't+1'. I have. e.g., two vectors:

y = [5678 6790 8904] %these are the latencies in trial t+1 in responding

x = [1 1 0] %these are the outcome of t, where 1 is rewarded and 0 is non rewarded

(I'm coming up with fake data, but just to give an example)

I'm doing now [b,bint,r,rint,stats] = regress(y,X) but notice some weird things:

1) not sure whether this is the right way of doing it

2) I have to put the vectors y' and x' otherwise it gives me error

3) for the 'stats' part it gives me NaN for F and p-value, does this mean it is not significant?

I also tried doing

table_latencies = table(y, x) table_latencies.x = nominal(table_latencies.x)

fit = fitlm(table_latencies,'y~x')

But I get this error:

Error using classreg.regr.FitObject/selectVariables (line 293) Predictor variables must be numeric vectors, numeric matrices, or categorical vectors.

Error in classreg.regr.TermsRegression/selectVariables (line 260) model = [e-mail](model);

Error in classreg.regr.FitObject/doFit (line 91) model = selectVariables(model);

Error in LinearModel.fit (line 1215) model = doFit(model);

Error in fitlm (line 117) model = LinearModel.fit(X,varargin{:});

Don't understand why, Thanks!!